• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.19 no.11
• /
• pp.94-101
• /
• 2020

In the age of industry 4.0, artificial intelligence is being widely used to realize machinery condition monitoring. Due to their excellent performance and the ability to handle large volumes of data, machine learning techniques have been applied to realize the fault diagnosis of different equipment. In this study, we performed the failure mode effect analysis (FMEA) of an aluminum electrolytic capacitor by using deep learning and big data. Several tests were performed to identify the main failure mode of the aluminum electrolytic capacitor, and it was noted that the capacitance reduced significantly over time due to overheating. To reflect the capacitance degradation behavior over time, we employed the Vanilla long short-term memory (LSTM) neural network architecture. The LSTM neural network has been demonstrated to achieve excellent long-term predictions. The prediction results and metrics of the LSTM and Vanilla LSTM models were examined and compared. The Vanilla LSTM outperformed the conventional LSTM in terms of the computational resources and time required to predict the capacitance degradation.

• International Journal of Advanced Culture Technology
• /
• v.11 no.4
• /
• pp.393-405
• /
• 2023

Accurate prediction of wind speed and power is vital for enhancing the efficiency of wind energy systems. Numerous solutions have been implemented to date, demonstrating their potential to improve forecasting. Among these, deep learning is perceived as a revolutionary approach in the field. However, despite their effectiveness, the noise present in the collected data remains a significant challenge. This noise has the potential to diminish the performance of these algorithms, leading to inaccurate predictions. In response to this, this study explores a novel feature engineering approach. This approach involves altering the data input shape in both Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) and Autoregressive models for various forecasting horizons. The results reveal substantial enhancements in model resilience against noise resulting from step increases in data. The approach could achieve an impressive 83% accuracy in predicting unseen data up to the 24th steps. Furthermore, this method consistently provides high accuracy for short, mid, and long-term forecasts, outperforming the performance of individual models. These findings pave the way for further research on noise reduction strategies at different forecasting horizons through shape-wise feature engineering.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2005.09b
• /
• pp.1189-1192
• /
• 2005

Due to their slow degradation properties, hydrophobic organic contaminants in estuarine sediment have been a concern for risks to human health and aquatic organisms. Studies of fate and transport of these contaminants in estuaries are further complicated by the fact that hydrodynamics and sediment transport processes in these regions are complex, involving processes with various temporal and spatial scales. In order to simulate and quantify long-term attenuation of Polycyclic Aromatic Hydrocarbons (PAH) in the Elizabeth River, VA, we develop a modeling approach, which employs the U.S. Environmental Protection Agency's water quality model, WASP, and encompasses key physical and chemical processes that govern long-term fate and transport of PAHs in the river. In this box-model configuration, freshwater inflows mix with ocean saline water and tidally averaged dispersion coefficients are obtained by calibration using measured salinity data. Sediment core field data is used to estimate the net deposition/erosion rate, treating only either the gross resuspension or deposition rate as the calibration parameter. Once calibrated, the model simulates fate and transport PAHs following the loading input to the river in 1967, nearly 4 decades ago. Sediment PAH concentrations are simulated over 1967-2022 and model results for Year 2002 are compared with field data measured at various locations of the river during that year. Sediment concentrations for Year 2012 and 2022 are also projected for various remedial actions. Since all the model parameters are based on empirical field data, model predictions should reflect responses based on the assumptions that have been governing the fate and sediment transport for the past decades.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.40 no.7
• /
• pp.471-475
• /
• 2016

In the present study, temperature evaluations on long-term storage of radioactive waste produced in the process of isotope production were performed using two different methods. Three-dimensional analysis was carried out assuming a volumetric heat source, while two-dimensional studies were performed assuming a point source. The maximum temperature difference between the predictions of the volumetric and point source models was approximately $5^{\circ}C$. For the conceptual design level, a point source model may be suitable to obtain the overall temperature characteristics of different loading locations. For more detailed analysis, the model with the volumetric source may be applicable to optimize the loading pattern in order to obtain minimum temperatures.

• Journal of the Korea Concrete Institute
• /
• v.20 no.1
• /
• pp.43-50
• /
• 2008

This paper presents experimental and analytical results on the long-term behavior of the reinforced recycled aggregate concrete beams under sustained loading. In this experimental program, three beams with different conditions of aggregates replacement (natural aggregate 100%, recycled coarse aggregate 100%, recycled fine aggregate 50%) were subjected to the sustained flexural loading that was a half of the nominal flexural capacity over a period of 1 year. The beam were designed with net span of 2,000 mm and rectangular cross-section of 170 mm width and 170 mm effective depth. The beams were instrumented and monitored to observe the change in the long-term behavior due to creep and shrinkage of concrete under sustained loading. The predictions of long-term deflection by ACI code, Branson, Mayer, Neville, EMM and AEMM were compared with the experimental results. From the experimental results, the reinforced concrete beams with recycled aggregates showed the same performance as that of a beam with natural aggregate. The proposed method to predict the long-term deflections of reinforced recycled aggregate concrete beams gives a good estimation for experimental results.

• Economic and Environmental Geology
• /
• v.52 no.6
• /
• pp.637-645
• /
• 2019

Due to the transformation to the intelligent information society, the rapid change of our life and environment is expected. The Ministry of Science and ICT (MSIT) and the National Research Council of Science and Technology (NST) introduced a five-year government supported research institution's planning and evaluation based on the mid-to long-term perspective. This study collects international benchmarking information including industry, academia, and research fields by collecting mid- and long-term strategy reports from public research institutes, surveys by experts from abroad universities and research institutes, and analyzing overseas market information reports. The British Geological Survey (BGS), the U.S. Geological Survey (USGS) and the japanese geological survey related institutes (AIST-GSJ) plans for three-dimensional national geological information, predictions of geological environmental disasters, and development of important metals and material in the low carbon economic transformation and in the era of the Fourth Industrial Revolution. The mid- and long-term program emphasizes basic and public research on geological information through abroad experts survey such as the IPGP-CNRS etc. The market analysis of the mining automation and digital map sectors has been able to derive the fields in which the role of public research institutes by the market is expected such as data collection on land and in the air, mobile or three-dimensional information production, smooth/fast/real-time maps, custom map design, mapping support to various platforms, geological environmental risk assessment and disaster management information and maps.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.6
• /
• pp.1143-1150
• /
• 2023

The global climate crisis and the implementation of low-carbon policies have led to a growing interest in renewable energy and a growing number of related industries. Among them, solar power is attracting attention as a representative eco-friendly energy that does not deplete and does not emit pollutants or greenhouse gases. As a result, the supplement of solar power facility is increasing all over the world. However, solar power is easily affected by the environment such as geography and weather, so accurate solar power forecast is important for stable operation and efficient management. However, it is very hard to predict the exact amount of solar power using statistical methods. In addition, the conventional prediction methods have focused on only short- or long-term prediction, which causes to take long time to obtain various prediction models with different prediction horizons. Therefore, this study utilizes a many-to-many structure of a recurrent neural network (RNN) to integrate short-term and long-term predictions of solar power generation. We compare various RNN-based very short- and long-term prediction methods for solar power in terms of MSE and R² values.

• Proceedings of The Korean Society of Agricultural and Forest Meteorology Conference
• /
• 2011.11a
• /
• pp.20-27
• /
• 2011

A system for fine-resolution long-range weather forecast is introduced in this study. The system is basically consisted of a global-scale coupled general circulation model (CGCM) and Weather Research and Forecast (WRF) regional model. The system makes use of a data assimilation method in order to reduce the initial shock or drift that occurs at the beginning of coupling due to imbalance between model dynamics and observed initial condition. The long-range predictions are produced in the system based on a non-linear ensemble method. At the same time, the model bias are eliminated by estimating the difference between hindcast model climate and observation. In this research, the predictability of the forecast system is studied, and it is illustrated that the system can be effectively used for the high resolution long-term weather prediction. Also, using the system, fine-resolution climatological data has been produced with high degree of accuracy. It is proved that the production of agrometeorological variables that are not intensively observed are also possible.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_4
• /
• pp.1125-1134
• /
• 2023

Out of the total 17,000 reservoirs in Korea, 13,600 small agricultural reservoirs do not have hydrological measurement facilities, making it difficult to predict water storage volume and appropriate operation. This paper examined univariate and multivariate long short-term memory (LSTM) modeling to predict the storage rate of agricultural reservoirs using remote sensing and artificial intelligence. The univariate LSTM model used only water storage rate as an explanatory variable, and the multivariate LSTM model added n-day accumulative precipitation and date of year (DOY) as explanatory variables. They were trained using eight years data (2013 to 2020) for Idong Reservoir, and the predictions of the daily water storage in 2021 were validated for accuracy assessment. The univariate showed the root-mean square error (RMSE) of 1.04%, 2.52%, and 4.18% for the one, three, and five-day predictions. The multivariate model showed the RMSE 0.98%, 1.95%, and 2.76% for the one, three, and five-day predictions. In addition to the time-series storage rate, DOY and daily and 5-day cumulative precipitation variables were more significant than others for the daily model, which means that the temporal range of the impacts of precipitation on the everyday water storage rate was approximately five days.

• International Union of Geodesy and Geophysics Korean Journal of Geophysical Research
• /
• v.23 no.1
• /
• pp.1-11
• /
• 1995

Thermospheric neutral winds and temperatures have been collected from the ground-based Fabry-Perot interferometer (FPI) at Thule Air Base ($76.5^{\circ}N{\;}69.0^{\circ}W$), Greenland since 1985. The thermospheric observations are obtained by determining the Doppler characteristics f the [OI] 6300 ${\AA}$ emissions of atomic oxygen. The FPI operates routinely during the winter season, with a limitation in the observation by the existence of clouds. For this study, data acquired from 1985 to 1991 were analyzed. The neutral wind measurements from these long-term measurements are used to investigate the influence of solar cycle variation on the high-latitude thermospheric dynamics. These data provide experimental results of the geomagnetic polar cap are also compared with the predictions of two semiempirical models : the vector spherical harmonics (VSH) model of Killeen et al. (1987) and the horizontal wind model (HWM) of Hedin et al. (1991). The experimental results show a good positive correlation between solar activity and thermospheric wind speed over the geomagnetic polar cap. The calculated correlation coefficient indicates that an increase of 100 in F10.7 index corresponds to an increase in wind speed of about 100 m/s. The model predictions reveal similar trends of wind speed variation as a function of solar activity, with the VSH and HWM models tending to overestimate and underestimate the wind speed, respectively.