• Journal of the Korean earth science society
• /
• v.44 no.4
• /
• pp.255-263
• /
• 2023

Western coastal area of Chungnam, including Cheonsu Bay and Garorim Bay, has suffered from hot and cold extremes. In this study, the extreme sea surface temperature on the western coast of Chungnam was analyzed using the quantile regression method, which extracts the linear regression values in all quantiles. The regional MOHID (MOdelo HIDrodinâmico) model, with a high resolution on a 1/60° grid, was constructed to reproduce the extreme sea surface temperature. For future prediction, the SSP5-8.5 scenario data of the CMIP6 model were used to simulate sea surface temperature variability. Results showed that the extreme sea surface temperature of Cheonsu Bay in August 2017 was successfully simulated, and this extreme sea surface temperature had a significant negative correlation with the Pacific decadal variability index. As a result of future climate prediction, it was found that an average of 2.9℃ increased during the simulation period of 86 years in the Chungnam west coast and there was a seasonal difference (3.2℃ in summer, 2.4℃ in winter). These seasonal differences indicate an increase in the annual temperature range, suggesting that extreme events may occur more frequently in the future.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2008.11a
• /
• pp.434-437
• /
• 2008

In this study, we developed a fuel moisture variation prediction model on each day after precipitation during a spring forest fire exhibition period. For this research, we selected plots in pine forest on Sam-Chuck si and Dong-hae si in Kangwon do according to a forest density(low, mediate, high) and classified a surface woody fuel by a diameter.(below 0.6cm, $0.6{\sim}3cm$, $3{\sim}6cm$, and above 6cm). A validity of this model was verified by applying a fuel moisture variation after precipitation in this spring. In the result, $R^2$ was $0.76{\sim}0.92$. This model will be a useful for improvement of a forest fire danger rate forcast through a prediction a fule moisture in forest.

• Journal of the Korean Society of Safety
• /
• v.32 no.1
• /
• pp.27-32
• /
• 2017

The experimental design methodology was applied in the real scale coal-fired boiler to predict the various combustion properties according to the operating conditions and to assess the coal plant safety. Response surface method (RSM) was introduced as a design of experiment, and the database for RSM was provided with the numerical simulation of the coal-fired boiler. The three independent variables, high heating value of coal (HHV), overall stoichiometry excess air ratio (OST), and burner-side stoichiometry excess air ratio (BST), were set to characterize the cross section averaged NOx concentration and temperature distribution. The maximum NOx concentration was predicted accurately and mainly controlled by BST in the boiler. The parabola function was assumed for the zone averaged peak temperature distribution, and the prediction was in a fairly good agreement with the experiments except downstream. Also, the location of the peak temperature was compared with that of maximum NOx, which implies that thermal NOx formation is the main mechanism in the coal-fired boiler. These results promise the wide use of statistical models for the fast prediction and safety assessment.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.40 no.11
• /
• pp.943-948
• /
• 2016

In recent times, the requirement of a micro pattern on the surface of products has been increasing, and high precision in the fabrication of the pattern is required. Hence, in this study, dual micro patterns were fabricated on a cylindrical workpiece, and deburring was performed by magnetic abrasive deburring (MAD) process. A prediction model was developed, and the MAD process was optimized using the response surface method. When the predicted values were compared with the experimental results, the average prediction error was found to be approximately 7%. Experimental verification shows fabrication of high accuracy dual micro pattern and reliability of prediction model.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.18 no.1
• /
• pp.1-5
• /
• 2005

A new model of plasma etch process was constructed by using a radial basis function network (RBFN). This technique was applied to an etching of silicon carbide films in a NF$_3$ inductively coupled plasma. Experimental data to train RBFN were systematically collected by means of a 2$^4$ full factorial experiment. Appropriateness of prediction models was tested with test data consisted of 16 experiments not pertaining to the training data. Prediction performance was optimized with variations in three training factors, the number of pattern units, width of radial basis function, and initial weight distribution between the pattern and output layers. The etch responses to model were an etch rate and a surface roughness measured by atomic force microscopy. Optimized models had the root mean-squared errors of 26.1 nm/min and 0.103 nm for the etch rate and surface roughness, respectively. Compared to statistical regression models, RBFN models demonstrated an improvement of more than 20 % and 50 % for the etch rate and surface roughness, respectively. It is therefore expected that RBFN can be effectively used to construct prediction models of plasma processes.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.13 no.1
• /
• pp.321-339
• /
• 2021

Response Surface Method (RSM) has been widely used for flammable cloud size prediction as it can reduce computational intensity for further Explosion Risk Analysis (ERA) especially during the early design phase of offshore platforms. However, RSM encounters the overfitting problem under very limited simulations. In order to overcome the disadvantage of RSM, Bayesian Regularization Artificial Neural (BRANN)-based model has been recently developed and its robustness and efficiency have been widely verified. However, for ERA during the early design phase, there seems to be room to further reduce the computational intensity while ensuring the model's acceptable accuracy. This study aims to develop an integrated method, namely the combination of Center Composite Design (CCD) method with Bayesian Regularization Artificial Neural Network (BRANN), for flammable cloud size prediction. A case study with constant and transient leakages is conducted to illustrate the feasibility and advantage of this hybrid method. Additionally, the performance of CCD-BRANN is compared with that of RSM. It is concluded that the newly developed hybrid method is more robust and computational efficient for ERAs during early design phase.

• Nuclear Engineering and Technology
• /
• v.56 no.1
• /
• pp.48-55
• /
• 2024

The safety-related raw water system's strong operational condition supports the radiation defense and biological shield of nuclear plant containment structures. Gaps and failures in maintaining proper working condition of main equipment like pump were among the most common causes of unavailability of safety related raw water systems. We integrated the advanced data analytics tools to evaluate the maintenance records of water systems and gave special consideration to deficiencies related to pump. We utilized maintenance data over a three-and-a-half-year period to produce metrics like MTBF, MTTF, MTTR, and failure rate. The visual analytic platform using tableau identified the efficacy of maintenance & deficiency in the safety raw water systems. When the number of water quality violation was compared to the other O&M deficiencies, it was discovered that water quality violations account for roughly 15% of the system's deficiencies. The pumps were substantial contributors to the deficit. Pump availability was predicted and optimized with real time data using response surface method. The prediction model was significant with r-squared value of 0.98. This prediction model can be used to predict forth coming pump failures in nuclear plant.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.14 no.1
• /
• pp.217-229
• /
• 1994

A probabilistic method was employed to compare the prediction accuracy of axial and volumetric strains of Lade's double surface model with that of single surface model. Several experiments were conducted to examine the variabilities of soil parameters for two models using Back-ma river sand. Mean values and standard deviations of soil parameters obtained from experimental data were used for the evaluation of the uncertainty of analyzed strains by the first order approximation. It is shown that the variabilities of parameters in the single surface model are more consistent than those of the double surface model. However, in the accuracy of axial strain by probabilistic analysis, double surface model is more stable than single surface model. It is also shown that two models are excellent in view of the accuracy of the volumetric strain. The method given in this paper may be effectively utilized to estimate the constitutive model because other results of the comparison of two models coincide with those of this paper.

• Journal of Ocean Engineering and Technology
• /
• v.9 no.2
• /
• pp.112-122
• /
• 1995

In this work, prediction of fatigue life and fatigue crack growth are studied. 4th order polynominal function is presented to describe the crack growth behaviors from artifical pit of SM45C steel. Crack growth curves obtained from 4th order polyminal growth equations are in good agreement with experimental data The crack growth behaviors at arbitrary stress levels and investigated by the concept of elastic-plastic fracture mechanics using ${\Delta}J$. Fatigue life prediction are carried out by numerical integral method. Prediction lives obtained by proposed method in this study, is in good agreement with the experimental ones. Life prediction results calculated by using of ${\Delta}J$ better than those of ${\Delta}K$.

• Ocean and Polar Research
• /
• v.23 no.2
• /
• pp.189-194
• /
• 2001

A practical algorithm for prediction of the sea surface temperatures (SST)from the satellite remote sensing data is presented in this paper. The fluctuations of SST consist of deterministic normals and stochastic anomalies. Due to large thermal inertia of sea water, the SST anomalies can be modelled by autoregressive or Markov process, and its near future values can be predicted provided the recent values of SST are available. The actual SST is predicted by superposing the pre-known SST normals and the predicted SST anomalies. We applied this prediction algorithm to the NOAA AVHRR weekly SST data for 18 years (1981-1998) in the seas adjacent to Korea (115-$145^{\circ}E$, 20-$55^{\circ}N$). The algorithm is applicable not only for prediction of SST in near future but also for nowcast of SST in the cloud covered regions.