• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.7 no.3
• /
• pp.132-136
• /
• 2014

We introduce a new architecture of TSK-based fuzzy inference system. The proposed model used fuzzy c-means clustering method(FCM) for efficient disposal of data. The premise part of fuzzy rules don't assume any membership function such as triangular, gaussian, ellipsoidal because we construct the premise part of fuzzy rules using FCM. As a result, we can reduce to architecture of model. In this paper, we are able to use four types of polynomials as consequence part of fuzzy rules such as simplified, linear, quadratic, modified quadratic. Weighed Least Square Estimator are used to estimates the coefficients of polynomial. The proposed model is evaluated with the use of Boston housing data called Machine Learning dataset.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.23 no.1
• /
• pp.108-114
• /
• 2012

We propose a dispersive finite-difference time domain(FDTD) algorithm suitable for the electromagnetic analysis of the human body. In this work, the dispersion relation of the human body is modeled by a quadratic complex rational function(QCRF), which leads to an accurate and efficient FDTD algorithm. Coefficients(involved in QCRF) for various human tissues are extracted by applying a weighted least square method(WLSM), referred to as the complex-curve fitting technique. We also presents the FDTD formulation for the QCRF-based dispersive model in detail. The QCRFbased dispersive model is significantly accurate and its FDTD implementation is more efficient than the counterpart of the Cole-Cole model. Numerical examples are used to show the validity of the proposed FDTD algorithm.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.16 no.5
• /
• pp.29-39
• /
• 2012

The main objective is to predict the future degradation and maintenance budget for a suspension bridge system. Bayesian inference is applied to find the posterior probability density function of the source parameters (damage indices and serviceability), given ten years of maintenance data. The posterior distribution of the parameters is sampled using a Markov chain Monte Carlo method. The simulated risk prediction for decreased serviceability conditions are posterior distributions based on prior distribution and likelihood of data updated from annual maintenance tasks. Compared with conventional linear prediction model, the proposed quadratic model provides highly improved convergence and closeness to measured data in terms of serviceability, risky factors, and maintenance budget for bridge components, which allows forecasting a future performance and financial management of complex infrastructures based on the proposed quadratic stochastic regression model.

• 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.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.4
• /
• pp.602-610
• /
• 2018

The model predictive control is an effective method to optimize the current control input that predicts the current control state and the future error using the predictive model of the control system when the reference trajectory is known. Since the control input can not have a physically infinitely large value, a predictive controller design with constraints should be considered. In addition, the reference model $A_r$ and the weight matrices Q, R that determine the control performance of the predictive controller are not optimized as arbitrarily designated should be considered in the controller design. In this study, we construct a predictive controller of a mobile robot by transforming it into a quadratic programming problem with constraints, The control performance of the mobile robot can be improved by optimizing the control parameters of the predictive controller that determines the control performance of the mobile robot using genetic algorithm. Through the computer simulation, the superiority of the proposed method is confirmed by comparing with the existing method.

• Fire Science and Engineering
• /
• v.33 no.2
• /
• pp.47-55
• /
• 2019

The prediction performance of design fire curves was evaluated using a Fire dynamics simulator (FDS) for a solid fuel fire in a building space by comparing the results with experimental data. EDC 2-step mixing controlled combustion model was used in the FDS simulations and the previously suggested 2-stage design fire (TDF), Quadratic and Exponential design fire curves were used as the FDS inputs. The simulation results showed that smoke propagation in the building space was significantly affected by the design fire curves. The predictions of simulations using design fire curves for the experimental temperatures in the building space were reasonable, but the TDF was found to be the most acceptable for predicting temperature. The predictions with each design fire curve of species concentrations showed insufficient agreement with the experiments. This suggests that the combustion model used in this study was not optimized for the simulation of a solid fuel fire, and additional studies will be needed to examine the combustion model on the FDS prediction of solid fires.

• Journal of Navigation and Port Research
• /
• v.36 no.10
• /
• pp.787-793
• /
• 2012

A hull form design technique to enhance the wave-making resistance performance for a medium size high speed Ro-Pax ship was studied introducing an optimization method and an automatic hull form modification method. SQP(sequential quadratic programming) was applied as the optimization algorithm and the geometry of hull surface was represented and modified using the NURBS(Non-Uniform Rational B-Spline). The wave-making resistance performance as an objective function in the optimization procedure was evaluated using the Rankine source panel method in which nonlinearity of the free surface boundary conditions and the trim and sinkage of the ship was fully taken into account. Using the Ro-Pax ship as a base hull, the hull-form optimization method was applied to obtain the hull shape that produced the lower wave-making resistance. To verify the validity of the hull-form optimization method, the numerical results was compared with the model test results.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.10a
• /
• pp.246-249
• /
• 2007

In numerical analysis for phase change material, numerical values of thermodynamic properties such as temperature, pressure, specific volume, enthalpy and entropy are required. But the steam table or diagram itself cannot be used without modelling. In this study applicability of neural networks in modelling superheated vapor region of water was examined by comparing with the quadratic spline. neural network consists of an input layer with 2 nodes, two hidden layers and an output layer with 3 nodes. Quadratic spline interpoation method was also applied for comparison. Neural network model revealed smaller percentage error to quadratic spline interpolation. From these results, it is confirmed that the neural networks could be powerful method in modelling the superheated range of the steam table.

• International Commerce and Information Review
• /
• v.19 no.1
• /
• pp.235-256
• /
• 2017

This study is to analyse how trust and asset specificity among partner firms affect on performance of international joint venture(IJV). Especially, the analysis was mainly based on a quadratic model. While it assumes that the previous studies was based on linear model in the relationship between trust, asset specificity and the performance, this study proceeds a empirical analysis by setting up a hypothesis; it would be quadratic relationship between trust, asset specificity and performance which are based on social capital theory and transaction cost theory. The survey was held with 74 manufactures who were established as an IJV by Korean and foreign firms together. In the result of the empirical analysis, trust shows an inverted U-shaped relationship with IJV performance. Also, asset specificity shows the U-shaped relationship with IJV performance. The results suggest that it needs to control and maintain the trust level among the partners in order not to lose an appropriate control caused by too much trust. In order to minimize the cost generated by asset specificity and to transform it into positive impact, it needs a control and the operation of monitoring system on the opportunistic action of the partners. Furthermore, it needs to keep organizational flexibility and innovativeness to continuously develop new capabilities.

• The Journal of the Korea Contents Association
• /
• v.20 no.8
• /
• pp.685-691
• /
• 2020

Standard values for fetal bio index measurements should be applied differently depending on the past present and general characteristics of the target population. Therefore, we tried to predict the number of gestational week(GA) and analyze the differences by country based on the measurements of Korean fetal bio index. 480 fetal bio index measurements between 15~38 weeks of pregnancy using ultrasound were compared retrospectively with USA ad Japanese data. One Way ANOVA was used for the analysis of differences by country, and quadratic regression model was designed to predict the GA of fetal bio index in order to predict the standard pregnancy number of Korean fetuses(p<0.005). Mean difference in the 95% confidence interval is BPD was Korea and USA 0.17, Korea and Japan 0.11, AC was Korea and USA -0.35, Korea and Japan 0.42, FL was Korea and USA -0.18, Korea and Japan 0.14. Therefore, fetal bio index for GA predict is considered to be the standard of the fetal growth assessment by applying the country specific standard in consideration of differences between races.