• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.4
• /
• pp.149-158
• /
• 2023

As data utilization is becoming more important recently, the importance of data centers is also increasing. However, the data center is a problem in terms of environment and economy because it is a massive power-consuming facility that runs 24 hours a day. Recently, studies using deep learning techniques to reduce power used in data centers or servers or predict traffic have been conducted from various perspectives. However, the amount of traffic data processed by the server is anomalous, which makes it difficult to manage the server. In addition, many studies on dynamic server management techniques are still required. Therefore, in this paper, we propose a dynamic server management technique based on Long-Term Short Memory (LSTM), which is robust to time series data prediction. The proposed model allows servers to be managed more reliably and efficiently in the field environment than before, and reduces power used by servers more effectively. For verification of the proposed model, we collect transmission and reception traffic data from six of Wikipedia's data centers, and then analyze and experiment with statistical-based analysis on the relationship of each traffic data. Experimental results show that the proposed model is helpful for reliably and efficiently running servers.

• Journal of Biosystems Engineering
• /
• v.37 no.4
• /
• pp.258-264
• /
• 2012

Purpose: This review aims at introducing 3 modeling approaches classified into 3 categories based on the purpose (estimation or prediction), structure (linear or non-linear) and phase (steady-state or dynamic-state); 1) statistical approaches, 2) kinetic modeling and 3) mechanistic modeling. We hope that this review can be a useful guide in the model-based approach of calcium metabolism as well as illustrates an application of engineering tools in studying biosystems. Background: The meaning of biosystems has been expanded, including agricultural/food system as well as biological systems like genes, cells and metabolisms. This expansion has required a useful tool for assessing the biosystems and modeling has arisen as a method that satisfies the current inquiry. To suit for the flow of the era, examining the system which is a little bit far from the traditional biosystems may be interesting issue, which can enlarge our insights and provide new ideas for prospective biosystem-researches. Herein, calcium metabolic models reviewed as an example of application of modeling approaches into the biosystems. Review: Calcium is an essential nutrient widely involved in animal and human metabolism including bone mineralization and signaling pathways. For this reason, the calcium metabolic system has been studied in various research fields of academia and industries. To study calcium metabolism, model-based system analyses have been utilized according to the purpose, subject characteristics, metabolic sites of interest, and experimental design. Either individual metabolic pathways or a whole homeostasis has been modeled in a number of studies.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.20 no.1
• /
• pp.98-105
• /
• 2010

Squeal, a kind of self-excited vibration, is generated by the friction between the disc and the friction materials. It occurs at the ending stage of the braking process, and radiates and audible frequency range of 1 kHz to 10 kHz. Squeal is generated from unstability because of the coupling between the translation and rotation of the system. This instability is caused by the follower force and follower force is normal component of the friction force. In this paper modal analysis of wheel brake system was performed in order to predict the squeal phenomenon. It was shown that the prediction of system instability is possible by FEM. A finite element model of that brake system was made. Some parts of a real brake was selected and modeled. Modal analysis method performs analyses of each brake system component. Experimental modal analysis was performed for each brake components and experimental results were compared with analytical results from FEM. To predict the dynamic unstability of a whole system, the complex eigenvalue analysis for assembly modeling of components confirmed by modal analysis is performed. The finite element models of the disk brake assembly have been constructed, and the squeal noise problems have been solved by complex eigenvalue analysis. The complex eigenvalue analysis results compared with real train test.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.9 no.1
• /
• pp.198-204
• /
• 2001

The shot peening is largely used for a surface treatment of metallic components where small spherical pellets called shots are blasted onto the surface with velocities up to 100 m/s. This treatment leads to improvement of fatigue behavior due to the developed compressive residual stresses, and so it has gained widespread acceptance I the automobile and aerospace industries. The residual stress profile on surface layer depends on the parameters of shot peening, which are, shot velocity, shot diameter, coverage, impact angle, material properties etc. and the method to confirm this profile is the measurement by X-ray diffractometer only. Despite the importance to automobile ad aerospace industries, little attention has been devoted to the accurate modelling of the process. In this paper, the simulation technique is applied to predict the magnitude ad distribution of the residual stress and plastic deformation caused by shot peening with the help of the finite element analysis.

• Water for future
• /
• v.25 no.3
• /
• pp.105-113
• /
• 1992

The purpose of this paper is to introduce and to apply the artificial neural network theory to real hydrologic system for forecasting daily streamflows during flood periods. The hydrologic dynamic process of rainfall-runoff is identified by the iterated estimation of system parameters that are determined by adjusting the weights of the network according to the non-linear response characteristics which is formed the model. Back propagation algorithm of neural network model is applied for the estimation of system parameters with past daily rainfall and runoff series data, and streamflows are forecasted using the parameters. The forecasted results are analyzed by statistical methods for the comparison with the observed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.8
• /
• pp.1647-1652
• /
• 2009

Generally the neural network and the fuzzy compensative algorithm are applied to forecast the time series for power demand with a characteristic of non-linear dynamic system, but it has a few prediction errors relatively. It also makes long term forecast difficult for sensitivity on the initial condition. On this paper, we evaluate the chaotic characteristic of electrical power demand with analysis methods of qualitative and quantitative and perform a forecast simulation of electrical power demand in regular sequence, attractor reconstruction, time series forecast for multi dimension using Lyapunov exponent quantitatively. We compare simulated results with the previous method and verify that the purpose one being more practice and effective than it.

• International Journal of Aeronautical and Space Sciences
• /
• v.11 no.2
• /
• pp.145-153
• /
• 2010

In this work, the comprehensive structural dynamics codes including DYMORE and CAMRAD II are used to validate the higher harmonic control aeroacoustic rotor test (HART) II data in descending flight condition. A total of 16 finite elements along with 17 aerodynamic panels are used for the CAMRAD II analysis; whereas, in the DYMORE analysis, 10 finite elements with 31 equally-spaced aerodynamic panels are utilized. To improve the prediction capability of the DYMORE analysis, the finite state dynamic inflow model is upgraded with a free vortex wake model comprised of near shed wake and trailed tip vortices. The predicted results on aerodynamic loads and blade motions are correlated with the HART II measurement data for the baseline, minimum noise and minimum vibration cases. It is found that an improvement of solution, especially for blade vortex interaction airloads, is achieved with the free wake method employed in the DYMORE analysis. Overall, fair to good correlation is achieved for the test cases considered in this study.

• Journal of Astronomy and Space Sciences
• /
• v.17 no.1
• /
• pp.107-116
• /
• 2000

This paper presents the preliminary survey and simulation results of the prediction of Leonid stream's orbital motion. Based on the model survey on eject velocity and perturbation of meteoroid particles, a simulation program was developed and applied to orbital motion of Leonid stream. The Jones ejection distribution model was used to describe the particle's eject velocity and the orbital dynamic model includes perturbations of major planet's gravity. DE405 ephemeris file generated by Solar System Dynamics Group at Jet Propulsion Laboratory in NASA was used for the planet's ephemeris calculations. Solar radiation pressure were also considered in the simulation and 8th order Runge-Kutta algorithm was used a numerical integration method.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.25 no.10
• /
• pp.1263-1273
• /
• 1988

An adaptive nulling technique is presented to synthetically overcome the integrated problems associated with the conventional LMS array in the performances of jammer rejection, convergence rate, misadjustment, and reference signal generation. The proposed method is to remove the target signal from the array input and to eliminate the reference signal prior to minimization processing. The algorithm is constrained to the residue noise level in adaptive processor. Analysis shows effectiveness of the algorithm for coherent and/or incoherent interference rejection, wide dynamic range of convergence factor, rapid adaptation rate, and small mean square error. Simulation results confirm the theoretical prediction.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2004.08a
• /
• pp.297-306
• /
• 2004

This study investigated Austenite Grain Size (AGS) distribution in Low-Speed Round-Oval Rolling. Rolling experiments were done along with the AGS numerical modeling to characterize the final AGS distribution and its kinetics behavior. For bar rolling experiment, we utilized the pilot rolling mill, operating at 34 fixed rpm, at POSCO Technical Research Laboratories. To investigate the microstructural observation, the rigid-viscoplastic finite element analysis was combined with Hodgson's AGS evolution model. To consider the transient thermal history in the integrative AGS modeling, additivity rule was introduced. The integrated analysis revealed that static or meta-dynamic recrystallization is responsible for the AGS difference in the inner or outer region of rolled bar. Comparative study showed that the current AGS modeling approach can be used to model the overall AGS distribution in bar rolling processes. For more accurate AGS prediction, the AGS modeling method should be verified under the various rolling conditions such as different rolling speeds and different deformations.