• Journal of Korea Water Resources Association
• /
• v.51 no.6
• /
• pp.503-514
• /
• 2018

In recent, the hydrological regime of the Mekong river is changing drastically due to climate change and haphazard watershed development including dam construction. Information of hydrologic feature like streamflow of the Mekong river are required for water disaster prevention and sustainable water resources development in the river sharing countries. In this study, runoff simulations at the Kratie station of the lower Mekong river are performed using SWAT (Soil and Water Assessment Tool), a physics-based hydrologic model, and LSTM (Long Short-Term Memory), a data-driven deep learning algorithm. The SWAT model was set up based on globally-available database (topography: HydroSHED, landuse: GLCF-MODIS, soil: FAO-Soil map, rainfall: APHRODITE, etc) and then simulated daily discharge from 2003 to 2007. The LSTM was built using deep learning open-source library TensorFlow and the deep-layer neural networks of the LSTM were trained based merely on daily water level data of 10 upper stations of the Kratie during two periods: 2000~2002 and 2008~2014. Then, LSTM simulated daily discharge for 2003~2007 as in SWAT model. The simulation results show that Nash-Sutcliffe Efficiency (NSE) of each model were calculated at 0.9(SWAT) and 0.99(LSTM), respectively. In order to simply simulate hydrological time series of ungauged large watersheds, data-driven model like the LSTM method is more applicable than the physics-based hydrological model having complexity due to various database pressure because it is able to memorize the preceding time series sequences and reflect them to prediction.

• Water for future
• /
• v.25 no.1
• /
• pp.93-100
• /
• 1992

A study on the Derivation of the Unit Hydrograph using Multiple Regression Moe이. The purpose of this study is to deriver an optimal unit hydrograph suing the multiple regression model, particularly when only small amount of data is available. The presence of multicollinearity among the input data can cause serious oscillations in the derivation of the unit hydrograph. In this case, the oscillations in the unit hydrograph ordinate are eliminated by combining the data. The data used in this study are based upon the collection and arrangement of rainfall-runoff data(1977-1989) at the Soyang-river Dam site. When the matrix X is the rainfall series, the condition number and the reciprocal of the minimum eigenvalue of XTX are calculated by the Jacobi an method, and are compared with the oscillation in the unit hydrograph. The optimal unit hydrograph is derived by combining the numerous rainfall-runoff data. The conclusions are as follows; 1)The oscillations in the derived unit hydrograph are reduced by combining the data from each flood event. 2) The reciprocals of the minimum eigen\value of XTX, 1/k and the condition number CN are increased when the oscillations are active in the derived unit hydrograph. 3)The parameter estimates are validated by extending the model to the Soyang river Dam site with elimination of the autocorrelation in the disturbances. Finally, this paper illustrates the application of the multiple regression model to drive an optimal unit hydrograph dealing with the multicollinearity and the autocorrelation which cause some problems.

• The Journal of Bigdata
• /
• v.8 no.2
• /
• pp.189-200
• /
• 2023

At the beginning of the COVID-19 pandemic, Korea's wine market had shrunk as other countries. However, right after the pandemic, Korea's imported wine consumption had been increased 69.6%. Because of the ban on overseas travel, wine was consumed in the domestic market. And consumption of high-end wines were increased significantly due to revenge spending and home drinking. However, from 2022 Korea's wine market has begun to shrink sharply again. Therefore this study forecasts the size of imported wine market by 2032 to provide useful information to wine related business entities. KITA(Korea International Trade Association)'s 95 time-series data per quarter from Q1 of 2001 to Q3 of 2023 was utilized in this research. The accuracy of model was tested based on value of MAPE. And ARIMA model was chosen to forecast the size of market value and Winter's multiplicative model was used for the size of market volume. The result of ARIMA model for the value (MAPE=10.56%) shows that the size of market value in 2032 will be increased up to USD $1,023,619, CAGR=6.22% which is 101% bigger than its size of 2023. On the other hand, the volume of imported wine market (MAPE=10.56%) will be increased up to 64,691,329 tons, CAGR=-0.61% which is only 15.12% bigger than its size of 2023. The result implies that the value of Korea's wine market will continue to grow despite the recent decline. And the high-end wine market will account for most of the increase.

• Journal of Korean Society of Environmental Engineers
• /
• v.38 no.8
• /
• pp.444-451
• /
• 2016

Most research on membrane fouling models in the past are based on theoretical equations in lab-scale experiments. But these studies are barely suitable for applying on the full-scale spot where there is a sequential process such as filtration, backwash and drain. This study was conducted in submerged membrane system which being on operation auto sequentially and treating wastewater from G-water purification plant in Incheon. TMP had been designated as a fouling indicator in constant flux conditions. Total volume of inflow and SS concentration are independent variables as major operation parameters and time-series analysis and prediction of TMP were conducted. And similarity between simulated values and measured values was assessed. Final prediction model by using genetic algorithm was fully adaptable because simulated values expressed pulse-shape periodicity and increasing trend according to time at the same time. As results of twice validation, correlation coefficients between simulated and measured data were $r^2=0.721$, $r^2=0.928$, respectively. Although this study was conducted limited to data for summer season, the more amount of data, better reliability for prediction model can be obtained. If simulator for short range forecast can be developed and applied, TMP prediction technique will be a great help to energy efficient operation.

• Korean Chemical Engineering Research
• /
• v.58 no.4
• /
• pp.588-595
• /
• 2020

Prediction and control of nitrogen oxides (NO_x) emission is of great interest in industry due to stricter environmental regulations. Herein, we propose an artificial intelligence (AI)-based framework for prediction of NO_x emission. The framework includes pre-processing of data for training of neural networks and evaluation of the AI-based models. In this work, Long-Short-Term Memory (LSTM), one of the recurrent neural networks, was adopted to reflect the time series characteristics of NO_x emissions. A decision tree was used to determine a time window of LSTM prior to training of the network. The neural network was trained with operational data from a heating furnace. The optimal model was obtained by optimizing hyper-parameters. The LSTM model provided a reliable prediction of NO_x emission for both training and test data, showing an accuracy of 93% or more. The application of the proposed AI-based framework will provide new opportunities for predicting the emission of various air pollutants with time series characteristics.

• Journal of Korea Water Resources Association
• /
• v.50 no.6
• /
• pp.359-371
• /
• 2017

This study generated flood time series of ungauged catchments in the Andongdam catchment using a distributed rainfall-runoff model and data generation method, and extracted the peak flows of 50 catchments to investigate the effect of rainfall spatial variability on peak flow simulation. The model performance statistics for three gauged catchments were reasonable for all events. The flood time series of the 50 catchments were generated using distributed and mean rainfall time series as input. The distribution of the peak flow using the mean rainfall was similar or slightly different to that using the distributed rainfall when the distribution of the distributed rainfall was nearly uniform. However, the distribution of the peak flow using the mean rainfall was reduced significantly compared to that using the distributed rainfall when actual storms moved to the top or bottom of the study catchment, or the rainfall was randomly distributed. These cases were 35% of total number events. Therefore, the spatial variability of rainfall should be considered for flood simulation. In addition, the power law relationship estimated using the peak flow of gauged catchments cannot be used for estimating the peak flow of ungauged independent catchments due to latter's significant variation of the peak flow magnitude.

• Korean Journal of Construction Engineering and Management
• /
• v.11 no.5
• /
• pp.41-52
• /
• 2010

Even though REITs (Real Estate Investment Trusts) are listed on the stock market, REITs have characteristics that allow them to invest in real estate and financing for real estate development. Therefore REITs is related with stock market and construction business and real estate business. Using time-series analysis, this study analyzed REITs in relation to construction businesses, real estate businesses, and the stock market, and derived influence factor of REITs. We used the VAR (vector auto-regression) and the VECM (vector error correction model) for the time-series analysis. This study classified three steps in the analysis. First, we performed the time-series analysis between REITs and construction KOSPI(The Korea composite stock price index) and the result showed that construction KOSPI influenced REITs. Second, we analyzed the relationship between REITs and construction commencement area of the coincident construction composite index, office index and housing price index in real estate business indexes. REITs and the housing price index influence each other, although there is no causal relationship between them. Third, we analyzed the relationship between REITs and the construction permit area of the leading construction composite index. The construction permit area is influenced by REITs, although there is no causal relationship between these two indexes, REITs influenced the stock market and housing price indexes and the construction permit area of the leading composite index in construction businesses, but exerted a relatively small influence in construction starts coincident with the composite office indexes in this study.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.8
• /
• pp.1551-1559
• /
• 2015

Recently, the performance of smart devices is almost similar to that of the existing PCs, thus the users of smart devices can perform similar works such as messengers, SNSs(Social Network Services), smart banking, etc. originally performed in PC environment using smart devices. Although the development of smart devices has led to positive impacts, it has caused negative changes such as an increase in security threat aimed at mobile environment. Specifically, the threats of mobile devices, such as leaking private information, generating unfair billing and performing DDoS(Distributed Denial of Service) attacks has continuously increased. Over 80% of the mobile devices use android platform, thus, the number of damage caused by mobile malware in android platform is also increasing. In this paper, we propose android based malware detection mechanism using time-series analysis, which is one of statistical-based detection methods.We use auto-regressive moving-average model which is extracting accurate predictive values based on existing data among time-series model. We also use fast and exact malware detection method by extracting possible malware data through Z-Score. We validate the proposed methods through the experiment results.

• Journal of Convergence for Information Technology
• /
• v.11 no.11
• /
• pp.57-65
• /
• 2021

We propose a novel method to detect abnormal data of specific symptoms using deep learning in air pollution measurement system. Existing methods generally detect abnomal data by classifying data showing unusual patterns different from the existing time series data. However, these approaches have limitations in detecting specific symptoms. In this paper, we use DeepLab V3+ model mainly used for foreground segmentation of images, whose structure has been changed to handle one-dimensional data. Instead of images, the model receives time-series data from multiple sensors and can detect data showing specific symptoms. In addition, we improve model's performance by reducing the complexity of noisy form time series data by using 'piecewise aggregation approximation'. Through the experimental results, it can be confirmed that anomaly data detection can be performed successfully.

• Korean Journal of Remote Sensing
• /
• v.34 no.6_4
• /
• pp.1545-1553
• /
• 2018

Recent eruptive activity at Kīlauea Volcano started on at the end of April in 2018 showed rapid ground deflation between May and June in 2018. On summit area Halema'uma'u lava lake continued to drop at high speed and Kīlauea's summit continued to deflate. GPS receivers and electronic tiltmeters detected the surface deformation greater than 2 meters. We explored the time-series surface deformation at Kīlauea Volcano, focusing on the early stage of eruptive activity, using multi-temporal COSMO-SkyMed SAR imagery. The observed maximum deformation in line-of-sight (LOS) direction was about -1.5 meter, and it indicates approximately -1.9 meter in subsiding direction by applying incidence angle. The results showed that summit began to deflate just after the event started and most of deformation occurred between early May and the end of June. Moreover, we confirmed that summit's deflation rarely happened since July 2018, which means volcanic activity entered a stable stage. The best-fit magma source model based on time-series surface deformation demonstrated that magma chambers were lying at depths between 2-3 km, and it showed a deepening trend in time. Along with the change of source depth, the center of each magma model moved toward the southwest according to the time. These results have a potential risk of including bias coming from single track observation. Therefore, to complement the initial results, we need to generate precise magma source model based on three-dimensional measurements in further research.