• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2008.04a
• /
• pp.299-305
• /
• 2008

In a series of research papers, the authors have studied some of the most important models of a contemporary universities, such as: the research university, the entrepreneurial university and the digital university and construct their Generalized Net (GN) models. This paper is based on the case-studies of Sofia University, the Technical University of Munich and the University of Edinburgh. The main focus is to put the analysis of the processes of the functioning of a university which effectively integrates Information and Communication Technologies (ICT) in all university activities. A concrete example based on the process of course administration at University of Edinburgh is considered. This university is in a process of developing an integrated information system covering most of the university activities. The opportunity of using GNs as a tool for modeling such processes is analyzed as well.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.364-369
• /
• 2023

Time-delay estimation between two receivers is a technique that has been applied in a variety of fields, from underwater acoustics to room acoustics and robotics. There are two types of time delay estimation techniques: one that estimates the amount of time delay from the correlation between receivers, and the other that parametrically models the time delay between receivers and estimates the parameters by system recognition. The latter has the characteristic that only a small fraction of the system's parameters are directly related to the delay. This characteristic can be exploited to improve the accuracy of the estimation by methods such as Lasso regularization. However, in the case of Lasso regularization, the necessary information is lost. In this paper, we propose a method using Elastic Net that adds Ridge regularization to Lasso regularization to compensate for this. Comparing the proposed method with the conventional Generalized Cross Correlation (GCC) method and the method using Lasso regularization, we show that the estimation variance is very small even for white Gaussian signal sources and colored signal sources.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.143-143
• /
• 2000

A neural net based generalized predictive control(NNGPC) is presented for a hydraulic servo position control system. The proposed scheme employs generalized predictive control, where the future output being generated from the output of artificial neural networks. The proposed NNGPC does not require an accurate mathematical model for the nonlinear hydraulic system and takes less calculation time than GPC algorithm if the teaming of neural network is done. Simulation studies have been conducted on the position control of a hydraulic motor to validate and illustrate the proposed method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.7
• /
• pp.1706-1725
• /
• 2024

The unique U-shaped structure of U-Net network makes it achieve good performance in image segmentation. This network is a lightweight network with a small number of parameters for small image segmentation datasets. However, when the medical image to be segmented contains a lot of detailed information, the segmentation results cannot fully meet the actual requirements. In order to achieve higher accuracy of medical image segmentation, a novel improved U-Net network architecture called multi-scale encoder-decoder U-Net+ (MEDU-Net+) is proposed in this paper. We design the GoogLeNet for achieving more information at the encoder of the proposed MEDU-Net+, and present the multi-scale feature extraction for fusing semantic information of different scales in the encoder and decoder. Meanwhile, we also introduce the layer-by-layer skip connection to connect the information of each layer, so that there is no need to encode the last layer and return the information. The proposed MEDU-Net+ divides the unknown depth network into each part of deconvolution layer to replace the direct connection of the encoder and decoder in U-Net. In addition, a new combined loss function is proposed to extract more edge information by combining the advantages of the generalized dice and the focal loss functions. Finally, we validate our proposed MEDU-Net+ MEDU-Net+ and other classic medical image segmentation networks on three medical image datasets. The experimental results show that our proposed MEDU-Net+ has prominent superior performance compared with other medical image segmentation networks.

• Nuclear Engineering and Technology
• /
• v.55 no.1
• /
• pp.285-294
• /
• 2023

Autonomous radiation source detection has long been studied for radiation emergencies. Compared to conventional data-driven or path planning methods, deep reinforcement learning shows a strong capacity in source detection while still lacking the generalized ability to the geometry in unknown environments. In this work, the detection task is decomposed into two subtasks: exploration and localization. A hierarchical control policy (HC) is proposed to perform the subtasks at different stages. The low-level controller learns how to execute the individual subtasks by deep reinforcement learning, and the high-level controller determines which subtasks should be executed at the current stage. In experimental tests under different geometrical conditions, HC achieves the best performance among the autonomous decision policies. The robustness and generalized ability of the hierarchy have been demonstrated.

• Journal of Environmental Science International
• /
• v.30 no.10
• /
• pp.811-820
• /
• 2021

The population variation of Spanish mackerel (Scomberomorus niphonius) according to its major prey abundance was analyzed using monthly catches of coastal set net fisheries in the southern waters off Gyeongsangnam-do and eastern waters off Gyeongsangbuk-do of Korea from 2006 to 2019. The abundance of Spanish mackerel and its prey species fluctuated almost simultaneously with time lags of +2 to -2 months between the set net fisheries in the southern and eastern waters. The generalized additive model revealed that the abundance of Spanish mackerel was influenced by its prey species such as hairtail and anchovy in southern waters, and common mackerel and horse mackerel in eastern waters. The model deviance explained 49% and 42% of Spanish mackerel abundance in southern and eastern waters respectively. These results suggest that the abundance of Spanish mackerel is affected by seasonal migratory prey fish species in the coastal areas and can be linked to their northerly migration.

• Nuclear Engineering and Technology
• /
• v.52 no.3
• /
• pp.461-468
• /
• 2020

The improvements of the DeCART/MUSAD code system for uncertainty analysis of HTGR neutronic parameters are presented in this paper. The function for quantifying an uncertainty of critical-spectrumweighted few group cross section was implemented using the generalized adjoint B₁ equation solver. Though the changes between the infinite and critical spectra cause a considerable difference in the contribution by the graphite scattering cross section, it does not significantly affect the total uncertainty. To reduce the number of iterations of the generalized adjoint transport equation solver, the generalized adjoint B₁ solution was used as the initial value for it and the number of iterations decreased to 50%. To reflect the implicit uncertainty, the correction factor was derived with the resonance integral. Moreover, an additional correction factor for the double heterogeneity was derived with the effective cross section of the DH region and it reduces the difference from the complete uncertainty. The code system was examined with the MHTGR-350 Ex.II-2 3D core benchmark. The k_eff uncertainty for Ex.II-2a with only the fresh fuel block was similar to that of the block and the uncertainty for Ex.II-2b with the fresh fuel and the burnt fuel blocks was smaller than that of the fresh fuel block.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.1
• /
• pp.188-194
• /
• 2015

This paper proposes a generalized regression neural network (GRNN) based algorithm for data interpolation and design optimization of brushless dc (BLDC) motor. The procedure makes use of magnet length, stator slot opening and air gap length as design variables. Cogging torque and average torque are treated as performance indices. The optimal design necessitates mitigating the cogging torque and maximizing the average torque by varying design variables. The data set for interpolation and ensuing design optimisation using GRNN is obtained by modeling a standard BLDC motor using finite element analysis (FEA) tool MagNet 7.1.1. The performance indices of the standard motor obtained using FEA are validated with an experimental model and an analytical method. The optimal design is authenticated using particle swarm optimization (PSO) algorithm and the performance indices of the optimal design obtained using GRNN is validated using FEA. The results indicate the suitability of GRNN as an interpolation and design optimization tool for a BLDC motor.

• Journal of Fisheries and Marine Sciences Education
• /
• v.28 no.1
• /
• pp.127-136
• /
• 2016

The purpose of this research is to conduct a comparative study on the characteristics of daily volatility spillovers across the stock markets of Korea, China, and Japan. We employ generalized spillover definition and measurement developed by Diebold & Yilmaz (2009, 2012). The sample period is January 5, 1993 to September 25, 2015. From a static full-sample analysis, we find that 8.60% of forecast error variance comes from volatility spillovers. From a 250-day rolling-sample analysis, we discover that there exist significant volatility fluctuations in the stock markets of Korea, China and Japan, expecially during the Asian Financial Crisis (1998-1999) and the US Credit Crisis (2008-2009) after the collapse of Lehman Brothers. From the net directional spillovers across three countries, we come upon that there is neither a definite leader nor a significant follower during the sample period.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.2932-2934
• /
• 1999

In this paper, a scheme for recognition of handwritten digits using a multilayer neural network trained with the back-propagation algorithm using generalized delta rule is proposed. The neural network is trained with hand written digit data of different writers and different styles. One of the purpose of the work with neural networks is the minimization of the mean square error(MSE) between actual output and desired one. The back-propagation algorithm is an efficient and very classical method. The back-propagation algorithm for training the weights in a multilayer net uses the steepest descent minimization procedure and the sigmoid threshold function. As an error rate is reduced, recognition rate is improved. Therefore we propose a method that is reduced an error rate.