• 한국데이터정보과학회:학술대회논문집
• /
• 2003.05a
• /
• pp.97-108
• /
• 2003

Cluster analysis has been widely used in many applications, such that pattern analysis or recognition, data analysis, image processing, market research on on-line or off-line and so on. Clustering can identify dense and sparse regions among data attributes or object attributes. But it requires many hours to get clusters that we want, because of clustering is more primitive, explorative and we make many data an object of cluster analysis. In this paper we propose a new method of clustering using sample based on grid. It is more fast than any traditional clustering method and maintains its accuracy. It reduces running time by using grid-based sample. And other clustering applications can be more effective by using this methods with its original methods.

• Journal of Korean Society for Quality Management
• /
• v.49 no.2
• /
• pp.201-211
• /
• 2021

Purpose: This paper proposes a density adaptive grid algorithm for the k-NN regression model to reduce the computation time for large datasets without significant prediction accuracy loss. Methods: The proposed method utilizes the concept of the grid with centroid to reduce the number of reference data points so that the required computation time is much reduced. Since the grid generation process in this paper is based on quantiles of original variables, the proposed method can fully reflect the density information of the original reference data set. Results: Using five real-life datasets, the proposed k-NN regression model is compared with the original k-NN regression model. The results show that the proposed density adaptive grid-based k-NN regression model is superior to the original k-NN regression in terms of data reduction ratio and time efficiency ratio, and provides a similar prediction error if the appropriate number of grids is selected. Conclusion: The proposed density adaptive grid algorithm for the k-NN regression model is a simple and effective model which can help avoid a large loss of prediction accuracy with faster execution speed and fewer memory requirements during the testing phase.

• Journal of the Society of Naval Architects of Korea
• /
• v.52 no.4
• /
• pp.338-348
• /
• 2015

In this study, the added resistance of ships in short waves is systematically studied by using two different numerical methods - Rankine panel method and Cartesian grid method – and existing asymptotic and empirical formulae. Analysis of added resistance in short waves has been preconceived as a shortcoming of numerical computation. This study aims to observe such preconception by comparing the computational results, particularly based on two representative three-dimensional methods, and with the existing formulae and experimental data. In the Rankine panel method, a near-field method based on direct pressure integration is adopted. In the Cartesian grid method, the wave-body interaction problem is considered as a multiphase problem, and volume fraction functions are defined in order to identify each phase in a Cartesian grid. The computational results of added resistance in short waves using the two methods are systematically compared with experimental data for several ship models, including S175 containership, KVLCC2 and Series 60 hulls (C_B = 0.7, 0.8). The present study includes the comparison with the established asymptotic and empirical formulae in short waves.

• The KIPS Transactions:PartC
• /
• v.11C no.6 s.95
• /
• pp.821-828
• /
• 2004

To analyze the usage information of system and network resources to the each grid application by measuring the real time traffic and calculating the statistic information, we suggested the kernel-based monitoring methods by researching the efficient monitoring method. This method use small system resourcesand measure the monitoring information accurately with less delay than the usual packet capture methods such as tcpdump. Also we implemented the monitoring systems which can monitor the used resources of system and network for grid application using the suggested kernel-based monitoring method. This research can give the useful information to the development of grid application and to grid network scheduler which can assign the proper resources to the grid application to perform efficiently. Network administrator can decide whether the expansion of network is required or not using the monitoring information.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.65 no.3
• /
• pp.69-82
• /
• 2023

Representative meteorological data of the rural water district, which is the spatial unit of the study, was produced using the grid-based national standard RCP scenario rainfall data provided by the Korea Meteorological Administration. The retrospective reproducibility of the climate model scenario data was analyzed, and the change in climate characteristics in the water district unit for the future period was presented. Finally the data characteristics and differences of each meteorological element according to various spatial resolution conversion and post-processing methods were examined. As a main result, overall, the distribution of average precipitation and R95p of the grid data, has reasonable reproducibility compared to the ASOS observation, but the maximum daily rainfall tends to be distributed low nationwide. The number of rainfall days tends to be higher than the station-based observation, and this is because the grid data is generally calculated using the area average concept of representative rainfall data for each grid. In addition, in the case of coastal regions, there is a problem that administrative districts of islands and rural water districts do not match. and In the case of water districts that include mountainous areas, such as Jeju, there was a large difference in the results depending on whether or not high rainfall in the mountainous areas was reflected. The results of this study are expected to be used as foundation for selecting data processing methods when constructing future meteorological data for rural water districts for future agricutural water management plans and climate change vulnerability assessments.

• Journal of Communications and Networks
• /
• v.14 no.6
• /
• pp.672-681
• /
• 2012

A potential breakthrough of the electrification of the vehicle fleet will incur a steep rise in the load on the electrical power grid. To avoid huge grid investments, coordinated charging of those vehicles is a must. In this paper, we assess algorithms to schedule charging of plug-in (hybrid) electric vehicles as to minimize the additional peak load they might cause. We first introduce two approaches, one based on a classical optimization approach using quadratic programming, and a second one, market based coordination, which is a multi-agent system that uses bidding on a virtual market to reach an equilibrium price that matches demand and supply. We benchmark these two methods against each other, as well as to a baseline scenario of uncontrolled charging. Our simulation results covering a residential area with 63 households show that controlled charging reduces peak load, load variability, and deviations from the nominal grid voltage.

• Journal of Power Electronics
• /
• v.19 no.2
• /
• pp.509-518
• /
• 2019

A novel islanding detection method is proposed in this paper. It is based on a frequency drooping PLL, which was presented in a previous work. The cause of errors in the non-detection zone (NDZ) of conventional frequency disturbance islanding detection methods (IDM) is analyzed. A frequency drooping phase-locked-loop (FD-PLL) is introduced into a single-phase grid-connected inverter (SPGCI), which can guarantee that grid current is in phase with the grid voltage. A novel FD-PLL IDM is proposed by improving this PLL. In order to verify the performance of the proposed FD-PLL IDM, a full performance comparison between the proposed IDM and typical existing active frequency drift IDMs is carried out, which includes both dynamic performance and steady performance. With the same NDZ, the total harmonic distortion of the grid-current in the dynamic process and steady state is analyzed. The proposed FD-PLL IDM, regardless of the dynamic or steady process, has the best power quality. Experimental and simulation results verify that the proposed FD-PLL IDM has excellent performance.

• Communications for Statistical Applications and Methods
• /
• v.29 no.1
• /
• pp.65-83
• /
• 2022

Although various statistical methods have been developed to map time-dependent genetic factors, most identified genetic variants can explain only a small portion of the estimated genetic variation in longitudinal traits. Gene-gene and gene-time/environment interactions are known to be important putative sources of the missing heritability. However, mapping epistatic gene-gene interactions is extremely difficult due to the very large parameter spaces for models containing such interactions. In this paper, we develop a Gaussian process (GP) based nonparametric Bayesian variable selection method for longitudinal data. It maps multiple genetic markers without restricting to pairwise interactions. Rather than modeling each main and interaction term explicitly, the GP model measures the importance of each marker, regardless of whether it is mostly due to a main effect or some interaction effect(s), via an unspecified function. To improve the flexibility of the GP model, we propose a novel grid-based method for the within-subject dependence structure. The proposed method can accurately approximate complex covariance structures. The dimension of the covariance matrix depends only on the number of fixed grid points although each subject may have different numbers of measurements at different time points. The deviance information criterion (DIC) and the Bayesian predictive information criterion (BPIC) are proposed for selecting an optimal number of grid points. To efficiently draw posterior samples, we combine a hybrid Monte Carlo method with a partially collapsed Gibbs (PCG) sampler. We apply the proposed GP model to a mouse dataset on age-related body weight.

• Journal of Information Processing Systems
• /
• v.15 no.5
• /
• pp.1156-1170
• /
• 2019

In recent years, the problem of data drifted of the smart grid due to manual operation has been widely studied by researchers in the related domain areas. It has become an important research topic to effectively and reliably find the reasonable data needed in the Supervisory Control and Data Acquisition (SCADA) system has become an important research topic. This paper analyzes the data composition of the smart grid, and explains the power model in two smart grid applications, followed by an analysis on the application of each parameter in density-based spatial clustering of applications with noise (DBSCAN) algorithm. Then a comparison is carried out for the processing effects of the boxplot method, probability weight analysis method and DBSCAN clustering algorithm on the big data driven power grid. According to the comparison results, the performance of the DBSCAN algorithm outperforming other methods in processing effect. The experimental verification shows that the DBSCAN clustering algorithm can effectively screen the power grid data, thereby significantly improving the accuracy and reliability of the calculation result of the main grid's theoretical line loss.

• 한국전산유체공학회:학술대회논문집
• /
• 1998.11a
• /
• pp.42-47
• /
• 1998

The pressure-based methods are very popular in CFD because it requires less computer core memory compared to other coupled or density-based solvers. Currently structured-mesh methodology based on pressure-based algorithm is quite mature to apply to the practical problems. The unstructured mesh method needs much more computer memory than the structured-mesh method. However the pressure-based method utilizing the sequential approach does not require very large memory used for unstructured-mesh density-based solvers. The present study has developed the unstructured grid pressure-based method. Cell-centered finite volume method was selected due to robustness for imposing various boundary conditions and easy implementation of higher-order upwind scheme. The predictive capability of present method has validated against several benchmark problems.