• Proceedings of the Korean Nuclear Society Conference
• /
• 1997.05a
• /
• pp.542-547
• /
• 1997

In the SSC(Super System Code), the reactivity feedback models of the Doppler effect and fuel axial expansion were modified to evaluate the safety performance of the metal-fueled core. The core radial expansion model was developed and implemented into the code as well. The transient analyses have been performed by the modified SSC for UTOP, ULOHS, ULOF/LOHS, and UTOP/LOF/LOHS events for one of the core design options being considered. Analysis results shows that the reactivity feedbacks can provide an inherent shutdown capability in response to key anticipated events without scram. Development of other reactivity feedback models and validation of these models against experimental data would make the SSC suitable for the assessment of the metal-fueled core safety performance.

• The Korean Journal of Applied Statistics
• /
• v.24 no.2
• /
• pp.307-313
• /
• 2011

In this paper, we compare performance of three models. The Holt-Winters, FARIMA and ARGARCH models, are used in predicting internet traffic data for analysis of traffic characteristics. We first introduce the time series models and apply them to real traffic data to forecast. Finally, we examine which model is the most suitable for explaining the long memory, the characteristics of the traffic material, and compare the respective prediction performance of the models.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.11S
• /
• pp.3179-3189
• /
• 1999

In this paper, different methods of integrating heterogeneous network information systems into the Web are observed by comparing a cliented Java CGI model with a server-oriented CGI model. In addition, a load balanced(LB, for short) CGI model is proposed, which combines two models and decides its course of action depending on the load state of the web server, and compared with the other two models. Performance evaluation models for three models are also presented. The results of computer simulations indicate that the LB CGI model performs consistently well irrespective of system load or server performance.

• Nuclear Engineering and Technology
• /
• v.49 no.2
• /
• pp.387-394
• /
• 2017

Seismic probabilistic risk assessment (SPRA) requires a large number of simulations to evaluate the seismic vulnerability of structural and nonstructural components in nuclear power plants. The effect of structural modeling and analysis assumptions on dynamic analysis of 3D and simplified 2D stick models of auxiliary buildings and the attached nonstructural components is investigated. Dynamic characteristics and seismic performance of building models are also evaluated, as well as the computational accuracy of the models. The presented results provide a better understanding of the dynamic behavior and seismic performance of auxiliary buildings. The results also help to quantify the impact of uncertainties associated with modeling and analysis of simplified numerical models of structural and nonstructural components subjected to seismic shaking on the predicted seismic failure probabilities of these systems.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.5
• /
• pp.120-128
• /
• 1993

In this paper, an implementation of high-speed parallel processing system for neural network design on the multicomputer network is presented. Linear speedup expandability is increased by reducing the synchronization penalty and the communication overhead. Also, we presented the parallel processing models and their performance evaluation models for each of the parallization methods of the neural network. The results of the experiments for the character recognition of the neural network bases on the proposed system show that the proposed approach has the higher linear speedup expandability than the other systems. The proposed parallel processing models and the performance evaluation models could be used effectively for the design and the performance estimation of the neural network on the multicomputer network.

• Communications for Statistical Applications and Methods
• /
• v.19 no.6
• /
• pp.885-898
• /
• 2012

In this paper, we consider Bayesian approaches to partially linear models, in which a regression function is represented by a semiparametric additive form of a parametric linear regression function and a nonparametric regression function. We make a comparative study on the performance of widely used Bayesian partially linear models in terms of empirical analysis. Specifically, we deal with three Bayesian methods to estimate the nonparametric regression function, one method using Fourier series representation, the other method based on Gaussian process regression approach, and the third method based on the smoothness of the function and differencing. We compare the numerical performance of three methods by the root mean squared error(RMSE). For empirical analysis, we consider synthetic data with simulation studies and real data application by fitting each of them with three Bayesian methods and comparing the RMSEs.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.3
• /
• pp.77-81
• /
• 2020

In this paper, we create scaffold defect classification models using machine learning based data. We extract the characteristic from collected scaffold external images using USB camera. SVM, KNN, MLP algorithm of machine learning was using extracted features. Classification models of three type learned using train dataset. We created scaffold defect classification models using test dataset. We quantified the performance of defect classification models. We have confirmed that the SVM accuracy is 95%. So the best performance model is using SVM.

• Journal of Korean Society for Atmospheric Environment
• /
• v.6 no.2
• /
• pp.125-134
• /
• 1990

The purpose of this study is to assess the performance of Short-term atmospheric dispersion models --- ISCST, MPTER, VALLEY --- with terrain adjustment. The models are evaluated through correlation analysis, paired analysis and log-normal culmulative analysis between the measured and predicted concentrations in Samcheonpo area. The correlation coefficients between the measured and predicted concentrations turn out to be higher with terrain adjustment than those without terrain adjustment. In paired analysis, the mean differences and average absolute gross errors of concentrations do not change significantly with terrain adjustment. But the variances of the residuals become much smaller when the terrain is adjusted. Through the log-normal cumulative analysis, it is found that the terrain adjustment improve the prediction performance of MPTER and VALLEY, but do not affect significantly that of ISCST. Overall, it is concluded that the performance of short term atmospheric dispersion models improve when the terrain is considered in computation, especially in MPTER and VALLEY.

• Journal of information and communication convergence engineering
• /
• v.21 no.1
• /
• pp.82-89
• /
• 2023

Various machine-learning models may yield high predictive power for massive time series for time series prediction. However, these models are prone to instability in terms of computational cost because of the high dimensionality of the feature space and nonoptimized hyperparameter settings. Considering the potential risk that model training with a high-dimensional feature set can be time-consuming, we evaluate a feature-importance-based feature selection method to derive a tradeoff between predictive power and computational cost for time series prediction. We used two machine learning techniques for performance evaluation to generate prediction models from a retail sales dataset. First, we ranked the features using impurity- and Local Interpretable Model-agnostic Explanations (LIME) -based feature importance measures in the prediction models. Then, the recursive feature elimination method was applied to eliminate unimportant features sequentially. Consequently, we obtained a subset of features that could lead to reduced model training time while preserving acceptable model performance.

• Journal of Korea Artificial Intelligence Association
• /
• v.1 no.2
• /
• pp.21-25
• /
• 2023

The purpose of this study was to compare the performance using multiple regression models to predict the energy consumption of steel industry. Specific independent variables were selected in consideration of correlation among various attributes such as CO2 concentration, NSM, Week Status, Day of week, and Load Type, and preprocessing was performed to solve the multicollinearity problem. In data preprocessing, we evaluated linear and nonlinear relationships between each attribute through correlation analysis. In particular, we decided to select variables with high correlation and include appropriate variables in the final model to prevent multicollinearity problems. Among the many regression models learned, Boosted Decision Tree Regression showed the best predictive performance. Ensemble learning in this model was able to effectively learn complex patterns while preventing overfitting by combining multiple decision trees. Consequently, these predictive models are expected to provide important information for improving energy efficiency and management decision-making at steel industry. In the future, we plan to improve the performance of the model by collecting more data and extending variables, and the application of the model considering interactions with external factors will also be considered.