• Industrial Engineering and Management Systems
• /
• v.14 no.3
• /
• pp.318-324
• /
• 2015

This paper proposes a dynamic short-term load forecasting method that utilizes a new sequential learning algorithm based on Relevance Vector Machine (RVM). The method performs general optimization of weights and hyperparameters using the current relevance vectors and newly arriving data. By doing so, the proposed algorithm is trained with the most recent data. Consequently, it extends the RVM algorithm to real-time and nonstationary learning processes. The results of application of the proposed algorithm to prediction of electrical loads indicate that its accuracy is comparable to that of existing nonparametric learning algorithms. Further, the proposed model reduces computational complexity.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2007.05a
• /
• pp.1441-1444
• /
• 2007

It is now widely acknowledged that climate variability modifies the frequency spectrum of hydrological extreme events. Traditional hydrological frequency analysis methodologies are not devised to account for nonstationarity that arises due to variation in exogenous factors of the causal structure. We use Hierarchical Bayesian Analysis to consider the exogenous factors that can influence on the frequency of extreme floods. The sea surface temperatures, predicted GCM precipitation, climate indices and snow pack are considered as potential predictors of flood risk. The parameters of the model are estimated using a Markov Chain Monte Carlo (MCMC) algorithm. The predictors are compared in terms of the resulting posterior distributions of the parameters associated with estimated flood frequency distributions.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 1997.11a
• /
• pp.451-462
• /
• 1997

A 2-dimensional numerical analysis is presented for thermal-electron heat conduction effects upon the electron transport and the drain current-voltage characteristics of submicron GaAs MESFETs, based on the use of a nonstationary hydrodynamic transport model. It is shown that for submicron GaAs MESFETs, electron heat conduction effects are significant on their internal electronic properties and also drain current-voltage characteristics. Due to electron heat conduction effects, the electron energy is greatly one-djmensionalized over the entire device region. Also, the drain current decreases continuously with increasing thermal conductivity in the saturation region of large drain voltages above 1 V. However, the opposite trend is observed in the linear region of small drain voltages below 1 V. Accordingly, for a large thermal conductivity, negative differential resistance drain current characteristics are observed with a pronounced peak of current at the drain voltage of 1 V. On the contrary, for zero thermal conductivity, a Gunn oscillation characteristic is observed at drain voltages above 2 V under a zero gate bias condition.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.6
• /
• pp.616-628
• /
• 2000

In this paper, we present the modeling of the ozone prediction system using Neuro-Fuzzy approaches. The mechanism of ozone concentration is highly complex, nonlinear, and nonstationary, the modeling of ozone prediction system has many problems and the results of prediction is not a good performance so far. The Dynamic Polynomial Neural Network(DPNN) which employs a typical algorithm of GMDH(Group Method of Data Handling) is a useful method for data analysis, identification of nonlinear complex system, and prediction of a dynamical system. The structure of the final model is compact and the computation speed to produce an output is faster than other modeling methods. In addition to DPNN, this paper also includes a Fuzzy Logic Method for modeling of ozone prediction system. The results of each modeling method and the performance of ozone prediction are presented. The proposed method shows that the prediction to the ozone concentration based upon Neuro-Fuzzy approaches gives us a good performance for ozone prediction in high and low ozone concentration with the ability of superior data approximation and self organization.

• ETRI Journal
• /
• v.16 no.2
• /
• pp.27-41
• /
• 1994

A new mathematical model to predict the beam pointing instability of a nonconservative two-body satellite system in spinning injection mode has been developed by using Newton-Euler and projection methods. Since the on-axis and null axis of the omni antenna with toroidal pattern beam form a right angle, wobbling of the antenna on-axis is measured by determining the Euler angles which represent the orientation of the satellite's spin axis. Because of the complexity of the system which is a time varying, nonstationary, nonlinear dynamical system, a numerical method is used for the analysis. Computer simulation results present the effects of the mass distribution and internal mass motion on the antenna beam pointing.

• Proceedings of the IEEK Conference
• /
• 2007.07a
• /
• pp.457-458
• /
• 2007

This paper propose a method for representing distributed data of sensor networks. The proposed method is based on a general distributed regression framework that models sensor data by fitting a global function to each of the local measurements and explores the possible extensions of distribution regression by using complex signal representations. In order to reduce the amount of processed data and the required communication, the signal model has to be as compact as possible, with the ability to restore the arbitrary measurements. To achieve this requirement, data compression step is included, where the basis function set is changed to an overcomplete set. This have better advantages in case of nonstationary signal modeling than complete base representation.

• Environmental Engineering Research
• /
• v.10 no.2
• /
• pp.88-103
• /
• 2005

In this paper, data-driven modeling and multiresolution analysis (MRA) are applied for a full-scale wastewater treatment plant (WWTP). The proposed method is based on modeling by partial least squares (PLS) and multiscale monitoring by a generic dissimilarity measure (GDM), which is suitable for nonstationary and non-normal process monitoring such as a biological process. Case study in an industrial plant showed that the PLS model could give good modeling performance and analyze the dynamics of a complex plant and MRA was useful to detect and isolate various faults due to its multiscale nature. The proposed method enables us to show the underlying phenomena as well as to filter out unwanted and disturbing phenomena.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.3
• /
• pp.563-569
• /
• 2004

The formulas taking into account the fundamental features of a meteoric radio propagation are obtained. Numerical simulation analysis has shown complex space structure of a field. Time behavior of intensity are researched taking into account nonstationary model. It is shown, this behavior essentially depends on parameters of a meteor trail, and that there is large variety of time dependencies of the signal intensity at the single scattering. In particular, at appropriate parameters of a meteor underdense trail it is possible large duration meteor bursts with which usually refer to an overdense meteor propagation.

• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.4
• /
• pp.127-134
• /
• 1995

The dynamic characteristics of a system can be critically influenced by system uncertainty, so the dynamic system must be analyzed stochastically in consideration of system uncertainty. This study presents a generalized stochastic model of dynamic system subjected to bot external and parametric nonstationary stochastic input. And this stochastic system is analyzed by a new stochastic process closure method and moment equation method. The first moment equation is numerically evaluated by Runge-Kutta method. But the second moment equation is founded to constitute an infinite coupled set of differential equations, so this equations are numerically evaluated by cumulant neglect closure method and Runge-Kutta method. Finally the accuracy of the present method is verified by Monte Carlo simulation.

• Journal of Korean Society on Water Environment
• /
• v.29 no.2
• /
• pp.247-258
• /
• 2013

A grid-based configuration of Land Surface Models (LSMs) coupled with a climate model can be advantageous in impact assessment of climate change for a large scale area. We assessed the applicability of Common Land Model (CoLM) to runoff and land surface temperature (LST) simulations at the domain that encompasses the Nakdong river basin. To establish a high resolution model configuration of a $1km{\times}1km$ grid size, both surface boundary condition and atmospheric inputs from the observed weather data in 2009 were adjusted to the same resolution. The Leaf Area Index (LAI) was collected from MODerate esolution Imaging Spectroradiometer (MODIS) and the downward short wave flux was produced by a nonstationary multi-site weather state model. Compared with the observed runoffs at the stations on Nakdong river, simulated runoffs properly responded to rainfall. The spatial features and the seasonal variations of the domain fairly well were captured in the simulated LSTs as well. The monthly and seasonal trend of LST were described well compared to the observations, however, the monthly averaged simulated LST exceeded the observed up to $2^{\circ}C$ at the 24 stations. From the results of our study, it is shown that high resolution LSMs can be used to evaluate not only quantity but also quality of water resources as it can capture the geographical features of the area of interest and its rainfall-runoff response.