• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.2
• /
• pp.398-406
• /
• 2011

In this paper, we introduce a new architecture of PSO-based Polynomial Neural Networks (PNN) and discuss its comprehensive design methodology. The conventional PNN is based on a extended Group Method of Data Handling (GMDH) method, and utilized the polynomial order (viz. linear, quadratic, and modified quadratic) as well as the number of node inputs fixed (selected in advance by designer) at Polynomial Neurons located in each layer through a growth process of the network. Moreover it does not guarantee that the conventional PNN generated through learning results in the optimal network architecture. The PSO-based PNN results in a structurally optimized structure and comes with a higher level of flexibility that the one encountered in the conventional PNN. The PSO-based design procedure being applied at each layer of PNN leads to the selection of preferred PNs with specific local characteristics (such as the number of input variables, input variables, and the order of the polynomial) available within the PNN. In the sequel, two general optimization mechanisms of the PSO-based PNN are explored: the structural optimization is realized via PSO whereas in case of the parametric optimization we proceed with a standard least square method-based learning. To evaluate the performance of the PSO-based PNN, the model is experimented with using Gas furnace process data, and pH neutralization process data. For the characteristic analysis of the given entire data with non-linearity and the construction of efficient model, the given entire system data is partitioned into two type such as Division I(Training dataset and Testing dataset) and Division II(Training dataset, Validation dataset, and Testing dataset). A comparative analysis shows that the proposed PSO-based PNN is model with higher accuracy as well as more superb predictive capability than other intelligent models presented previously.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.53 no.6
• /
• pp.415-424
• /
• 2004

In this paper, we introduce and investigate a new category of rule-based fuzzy inference system based on Information Granulation(IG). The proposed rule-based fuzzy modeling implements system structure and parameter identification in the efficient form of “If..., then...” statements, and exploits the theory of system optimization and fuzzy implication rules. The form of the fuzzy rules comes with three types of fuzzy inferences: a simplified one that involves conclusions that are fixed numeric values, a linear one where the conclusion part is viewed as a linear function of inputs, and a regression polynomial one as the extended type of the linear one. By the nature of the rule-based fuzzy systems, these fuzzy models are geared toward capturing relationships between information granules. The form of the information granules themselves becomes an important design features of the fuzzy model. Information granulation with the aid of HCM(Hard C-Means) clustering algorithm hell)s determine the initial parameters of rule-based fuzzy model such as the initial apexes of the membership functions and the initial values of polynomial function being used in the Premise and consequence Part of the fuzzy rules. And then the initial Parameters are tuned (adjusted) effectively with the aid of the improved complex method(ICM) and the standard least square method(LSM). In the sequel, the ICM and LSM lead to fine-tuning of the parameters of premise membership functions and consequent polynomial functions in the rules of fuzzy model. An aggregate objective function with a weighting factor is proposed in order to achieve a balance between performance of the fuzzy model. Numerical examples are included to evaluate the performance of the proposed model. They are also contrasted with the performance of the fuzzy models existing in the literature.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.12 no.3
• /
• pp.9-20
• /
• 2009

The naturalized plants in the Gwangyang Steel Works were listed 70 taxa; 18 families, 51 genus, 66 species, 3 varieties and 1 form. by field survey. The naturalized plants divided into pattern by survey of annual plants ratio is 31 taxa (44.3%) by life form spectrum and perennials is 23 taxa (32.9%), biennials is 13 taxa (18.5%), two trees (Robinia pseudoacacia, Alianthus altissima) and one shrub (Amorpha furticosa) were founded. According to analysis results form place of origin, Europe covered 26 taxa (37.1%), North America covered 23 taxa (32.8%). Naturalized degree 3 plants, as common but not abundant, founded as 24 taxa (35.0%) were hold most highly ratio, naturalized degree 1 plants founded as 3 taxa (4.0%) were hold lowest. Introduction period 1 covered 31 taxa (44.3%) have had highest score and period 2 covered 11 taxa (15.7%) have had lowest scores. In addition, the urbanization index based on 271 taxa was 25.8% and 2 taxa (Solanum carolinense, Ambrosia artemisiaefolia) is growing in the Gwangyang Steel Works by ecosystem disturbing wild plants. Meanwhile, Ambrosia artemisiaefolia have confirmed into several tens~hundred in the Gwangyang Steel Works were damage the human body of plants develop an allergy to pollen. Solanium carolinense have composed several large communities about $10m{\times}10m$ ($100m^2$) and hold a dominant position, so management plan of the sequel of monitoring them might be required.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.7 no.1
• /
• pp.33-45
• /
• 1995

This paper as a sequel to Kang et al. (1994) describes the development of a three-dimensional nonlinear Galerkin-FEM model. Nonlinear advective terms have been incorporated in a manner used by Lardner and Song (1992), that is, using velocities at given nodes computed with linear Galerkin FEM model. The Proposed model is computationally more efficient than previous nonlinear Galerkin models developed by Owen (1980) and Davies (1980) because the model uses a linear shape function as a basis and, furthermore, the similarity transform technique developed by Kang (1994). Two experiments have been carried out to examine effects of nonlinear terms. One is an experiment of wind-driven current in a rectangular basin (Heaps' basin) and the other is an experiment concerning eddy generation behind a jetty with specified downstream and upstream open boundary conditions. The computed Pattern was found to be in good agreement qualitatively with previous model experiments by Stelling (1984).

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.31 no.1
• /
• pp.120-124
• /
• 2003

KAISTSAT- 4, an experimental small satellite, is being developd by Satellite Technology Research Center in KAIST as a sequel mission to KITSAT-1, 2, and 3. The flight model scheduled to be launched in 2003, the qualification model construction and testing have been completed recently. The satellite subsystems of the qualification model have been tested under a thermal vacuum environment harsher than expected in the orbit. Thermal balance test has also been done in order to evaluate and tune the thermal analysis model of the qualification model. This paper describes the thermal vacuum test procedure, the results, as well as the lessons learned during the tests, which can be useful for future thermal vacuum tests of small satellites.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.12
• /
• pp.1772-1781
• /
• 2017

In this study, the design methodology for alleviating the overfitting problem of Polynomial Neural Networks(PNN) is realized with the aid of two kinds techniques such as L2 regularization and Sum of Squared Coefficients (SSC). The PNN is widely used as a kind of mathematical modeling methods such as the identification of linear system by input/output data and the regression analysis modeling method for prediction problem. PNN is an algorithm that obtains preferred network structure by generating consecutive layers as well as nodes by using a multivariate polynomial subexpression. It has much fewer nodes and more flexible adaptability than existing neural network algorithms. However, such algorithms lead to overfitting problems due to noise sensitivity as well as excessive trainning while generation of successive network layers. To alleviate such overfitting problem and also effectively design its ensuing deep network structure, two techniques are introduced. That is we use the two techniques of both SSC(Sum of Squared Coefficients) and $L_2$ regularization for consecutive generation of each layer's nodes as well as each layer in order to construct the deep PNN structure. The technique of $L_2$ regularization is used for the minimum coefficient estimation by adding penalty term to cost function. $L_2$ regularization is a kind of representative methods of reducing the influence of noise by flattening the solution space and also lessening coefficient size. The technique for the SSC is implemented for the minimization of Sum of Squared Coefficients of polynomial instead of using the square of errors. In the sequel, the overfitting problem of the deep PNN structure is stabilized by the proposed method. This study leads to the possibility of deep network structure design as well as big data processing and also the superiority of the network performance through experiments is shown.

• Journal of Society of Preventive Korean Medicine
• /
• v.5 no.1
• /
• pp.90-102
• /
• 2001

Objectives: Resently Oxidative stress of brain was proved the cause of Alzheimer and stroke sequel. It has important significance in prevention and treatment of cerebropathia that Bulnohwan used as formula of senescence delay have antioxidative effect. The purposes of this study is to investigate the effect of Bulnohwan on antioxidant defense systems such as thiobarbituric acid reactive substances(TBARS), Superoxide dismutase (SOD), Catalase (CAT), Glutathione peroxidase (GSH-PX), Glutathione S-transperase (GST), Glutathione (GSH) in rat brain. Method: Sprague - Dawley rats were divided into 3 groups; saline solution administered control group, Bulnohwan extract administered Experimental group I and Bulnohwan adminisrtrated, 40% dietary restricted Experimental group II. Animals were sacrificed at 12 weeks after treatment TBARS, SOD, CAT, GSH-PX, GST and GSH were measured in mts brain. Results: weight of brain was no stastical significance.(p>0.05) TBARS contents were significant decrease in Experimental group I, II. (p<0.001) SOD activity was stastical significance in Experimental group II, whereas it was no stastical significance Experimental group II.(p<0.0001) Catalase activites were significant increase in . (p<0.00l) Glutathione Peroxidase activites were significant increase in Experimental group I,II. (p<0.000l) Glutathione S-transferase activites were significant increase in Experimental group I, II. (p<0.000) However there were no statistical significance each other. Glutathione contents were significant increase in Experimental group I, II. (p<0.00l) Conclusions: According to the above results, it is considered that Bulohwan has antioxidants effect in rat brain. When Bulohwan goes with diet restriction, there has more Antioxidants effect in rat brain. but this study was perfored retrospectively. So more prospective studies about mutual relation of drugs are needed

• International Area Studies Review
• /
• v.15 no.3
• /
• pp.191-213
• /
• 2011

This study empirically examines whether income smoothing effectively contributes to consumption smoothing. Since international factor movement plays an important role for income smoothing across countries, net factor income across countries deserves more attention in considering income smoothing than attention having been paid by previous studies in the literature. Most of previous studies assume that net factor income has the same degree of effects on consumption as GDP. For the case of 12 EU countries during the period from 1999 to 2010, our empirical investigation observes that the response of consumption with respect to net factor income is severely lower than the response of consumption with respect to GDP, and further net factor income has ignorable effects on the consumption. This evidence implies that the income smoothing via international factor movement is less significant in contributing to the consumption smoothing. In the sequel, our finding suggests that legal and institutional process for transferring net factor income across countries should be eased further to improve contribution of income smoothing to consumption smoothing.

• Ocean Systems Engineering
• /
• v.9 no.2
• /
• pp.191-218
• /
• 2019

The tower-platform interface and mooring system of floating offshore wind turbines (FOWTs) are some of the most critical components with significant influences on overall project costs. In addition to satisfying strength requirements, it is typical and vital to meet fatigue criteria for a service life of 25 years or more. Wind spectra characteristics considered in analysis can penalize fatigue designs, leading to unnecessary costs. The International Electrotechnical Commission (IEC, 2009) recommends the use of site-specific wind data (spectrum, turbulence intensity, etc.) in design of FOWTs, but for offshore sites it is often the case that such data is unavailable and land-based data are used as surrogates in design. For such scenarios, it is worth investigating whether such alternative approach is suitable and accurate, and understanding the consequence of the selection of wind spectral characteristics on fatigue design. This paper addresses the impact of the subsequent selection on fatigue responses of towerbase and mooring system in a FOWT, as a sequel to the paper by Udoh and Zou (2018) which focused on impacts on strength design. The 5 MW semi-submersible FOWT platform with six mooring lines implemented in the preceding study is applied in analysis. Results indicate significant variations in resulting fatigue life with considered wind parameters. Thus, it is critical to apply proper wind spectra characteristics for analysis and design of FOWTs to avoid unnecessary conservatism and costs. Based on the findings of this study, more explicit guidance on the application of turbulence intensities for IEC-recommended models in offshore sites could lead to more accurate load estimates in design of FOWTs.

• Tribology and Lubricants
• /
• v.38 no.3
• /
• pp.84-92
• /
• 2022

In Part I, developed was a method to obtain the stress field due to an edge dislocation that locates in an elastic half plane beneath the contact edge of an elastically similar square wedge. Essential result was the corrective functions which incorporate a traction free condition of the free surfaces. In the sequel to Part I, features of the corrective functions, F_kij,(k = x, y;i,j = x,y) are investigated in this Part II at first. It is found that F_xxx(ŷ) = F_xyx(ŷ) where ŷ = y/η and η being the location of an edge dislocation on the y axis. When compared with the corrective functions derived for the case of an edge dislocation at x = ξ, analogy is found when the indices of y and x are exchanged with each other as can be readily expected. The corrective functions are curve fitted by using the scatter data generated using a numerical technique. The algebraic form for the curve fitting is designed as F_kij(ŷ) = $\frac{1}{\hat{y}^{1-{\lambda}}I+yp}$$\sum_{q=0}^{m}{\left}$$\left[A_q\left(\frac{\hat{y}}{1+\hat{y}} \right)^q \right]$ where λ_I=0.5445, the eigenvalue of the adhesive complete contact problem introduced in Part I. To investigate the exponent of F_kij, i.e.(1 - λ_I) and p, Log|F_kij|(ŷ)-Log|(ŷ)| is plotted and investigated. All the coefficients and powers in the algebraic form of the corrective functions are obtained using Mathematica. Method of analyzing a surface perpendicular crack emanated from the complete contact edge is explained as an application of the curve-fitted corrective functions.