• ETRI Journal
• /
• v.37 no.2
• /
• pp.283-294
• /
• 2015

A model that precisely forecasts how much wind power is generated is critical for making decisions on power generation and infrastructure updates. Existing studies have estimated wind power from wind speed using forecasting models such as ANFIS, SMO, k-NN, and ANN. This study applies a projected clustering technique to identify wind power patterns of wind turbines; profiles the resulting characteristics; and defines hourly and daily power patterns using wind power data collected over a year-long period. A wind power pattern prediction stage uses a time interval feature that is essential for producing representative patterns through a projected clustering technique along with the existing temperature and wind direction from the classifier input. During this stage, this feature is applied to the wind speed, which is the most significant input of a forecasting model. As the test results show, nine hourly power patterns and seven daily power patterns are produced with respect to the Korean wind turbines used in this study. As a result of forecasting the hourly and daily power patterns using the temperature, wind direction, and time interval features for the wind speed, the ANFIS and SMO models show an excellent performance.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.4
• /
• pp.293-301
• /
• 2001

This paper reviews three methods to compute a pointwise confidence interval of neuro-fuzzy models and compares their estimation perfonnanee through simulations. The eOITl.putation methods under consideration include stacked generalization using cross-validation, predictive error bar in regressive models, and local reliability measure for the networks employing a local representation scheme. These methods implemented on the neuro-fuzzy models are applied to the problems of simple function approximation and chaotic time series prediction. The results of reliability estimation are compared both quantitatively and qualitatively.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2007.11a
• /
• pp.1442-1445
• /
• 2007

Just noticeable differences (jnds) of autocorrelation function (ACF) parameters - Phi (0), Tau 1 and Phi 1 - of household refrigerator noise were investigated by psychoacoustical analyses. Phi (0) of five refrigerators' noise was changed with equal (${\pm}$) interval level of 0.5-1.0 dB up to five intervals by manipulating sound pressure level of the noise. Tau 1 and Phi 1 were varied with equal (${\pm}$) interval of around 0.10 ms and 0.02, respectively. Pitch shifting and strengthen methods were applied for the Tau 1 and Phi 1 variations. As results of subjective evaluations, about 2.0 dB was shown as jnd of Phi (0). The values of 0.30 ms and 0.06 were found as jnds of Tau 1 and Phi 1, respectively. The jnd results of each ACF parameter can be applied to explain substantial amount of sound quality (SQ) enhancement in the SQ prediction indices which were proposed in the authors' previous study [Sato et al. (2007) J. Acoust. Soc. Am.].

• The KIPS Transactions:PartD
• /
• v.8D no.4
• /
• pp.387-392
• /
• 2001

This Many software projects collect grouped failure data (failures in some failure interval or in variable time interval) rather than individual failure times or failure count data during the testing or operational phase. This paper presents the neural network (NN) modeling for grouped failure data that is able to predict cumulative failures in the variable future time. The two variant models of cascade-correlation learning (CasCor) algorithm are presented. Suggested models are compared with other well-known NN models and statistical software reliability growth models (SRGMs). Experimental results show that the suggested models show better predictability.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.13 no.3
• /
• pp.77-82
• /
• 2014

In this study, we carried out the interrupted cutting of carbon steel for a machine structure (SM45C) with a CVD-coated tool and conducted an ANOVA test and a confidence interval analysis to find factors influence the surface roughness and to obtain a regression equation. We found that factor which mostly affects the surface roughness during interrupted cutting was the feed rate. The cutting speed and depth of the cut only had small effect on the surface roughness. From the result of a multi-regression analysis during an interrupted cutting experiment, we obtained regression equation. Its coefficient of determination was 0.918, indicating that the regression equation was predictable. Compared to continuous cutting, if the feed rate increases, the surface roughness will also increase during interrupted cutting.

• Journal of Environmental Science International
• /
• v.21 no.5
• /
• pp.613-622
• /
• 2012

A pilot-scale pulse-jet bagfilter was designed, built and tested for the effects of four operating conditions (filtration velocity, inlet dust concentration, pulse pressure, and pulse interval time) on the total system pressure drop, using coke dust from a steel mill factory. Two models were used to predict the total pressure drop according to the operating conditions. These model parameters were estimated from the 180 experimental data points. The empirical model (EM) with filtration velocity, areal density, inlet dust concentration, pulse interval time and pulse pressure shows the best correlation coefficient (R=0.971) between experimental data and model predictions. The empirical model was used as it showed higher correlation coefficient (R=0.971) compared to that of the Multivariate linear regression(MLR) (R=0.961). The minimum pulse pressure predicted by empirical model (EM) was 5kg/$cm^2$.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.55 no.11
• /
• pp.495-501
• /
• 2006

Diagnostics of unborn baby is mainly aimed at prediction and detection of occurrence of intrauterine hypoxia. Consequences resulting from fetal hypoxia appear in its heart activity. In this study, we have developed a new non-invasive system for fetal hypoxia diagnosis which provides systolic time interval(STI) parameters on the basis of analysis of electrical and mechanical heart activity together. For this we have worked on 1) the proper lead system for the acquisition of abdominal ECG, 2) the independent component analysis based signal processing and fetal ECG separation, 3) the development of a hardware which consists of an abdominal ECG amplifying module and an ultrasound module and 4) the detection of characteristic points of FECG and Doppler signal and the extraction of diagnostic parameters. The developed system was evaluated by the clinical experiments in which 33 subjects were participated. The acquired STI by the system were distributed within the ranges from the well-established invasive results of other researchers. From this, we can conclude that the developed non-invasive fetal hypoxia diagnosis system is useful.

• KIISE Transactions on Computing Practices
• /
• v.22 no.5
• /
• pp.235-239
• /
• 2016

'Commit-bug link', the link between commit history and bug reports, is used for software maintenance and defect prediction in bug tracking systems. Previous studies have shown that the links are automatically detected based on text similarity, time interval, and keyword. Existing approaches depend on the quality of commit history and could thus miss several links. In this paper, we proposed a technique to link commit and bug report using not only messages of commit history, but also the similarity of files in the commit history coupled with bug reports. The experimental results demonstrated the applicability of the suggested approach.

• Journal of the Korean Burn Society
• /
• v.22 no.2
• /
• pp.38-44
• /
• 2019

Purpose: The immature granulocyte count has been reported to be a marker of infection and sepsis. The difference in leukocyte subfractions (delta neutrophil index, DNI) in ADVIA 2120 reflects the fraction of circulating immature granulocytes in the blood. This study evaluated the clinical utility of DNI as a severity and prediction marker in critically ill patients with burn sepsis. Methods: One hundred and sixty nine patients admitted to the burn care unit were studied. DNI (the difference in leukocyte subfractions identified by myeloperoxidase and nuclear lobularity channels) was determined using a specific blood cell analyzer. Results: Seventy one patients (42 %) were diagnosed with burn sepsis. DNI was significantly higher in patients with burn sepsis than in patients without (P<0.01). Delta neutrophil index was a better indicator of burn sepsis than C-reactive protein, lactate, white blood cell count, HCO3, base excess, lactate, procalcitonin (odds ratio, 6.31; confidence interval 2.36~16.90; P<0.01). And the receiver operating characteristic curves showed that delta neutrophil index, AUC 0.795 (95% confidence interval, 0.721~0.869; P<0.05) was a better predictor of burn sepsis than lactate, procalcitonin, white blood cell, base excess and abbreviated burn severity index. Conclusion: Delta neutrophil index may be used as a early marker of patients with burn sepsis.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.7A
• /
• pp.762-769
• /
• 2007

Adaptive MIMO-OFDM (Orthogonal Frequency Division Multiplexing) system adaptively changes modulation scheme depending on feedback channel state information (CSI). The CSI feedback channel which is the reverse link channel has multiple symbol delays including propagation delay, processing delay, frame delay, etc. The unreliable CSI due to feedback delay degrades adaptive modulation system performance. This paper compares the MSE and data capacity with respect to delay and channel signal to noise ratio for the two multi-step channel prediction schemes, CTSBP and BTSBP, such that robust adaptive SISO-OFDM/MIMO-OFDM is designed over severe mobile multipath channel conditions. This paper presents an interpolation method to reduce feedback overhead for adaptive MIMO-OFDM and shows MSE with respect to interpolation interval.