• Journal of the Korean Society of Marine Environment & Safety
• /
• v.26 no.4
• /
• pp.382-391
• /
• 2020

In this study, we used artificial intelligence algorithms for the prediction of dissolved oxygen in Jindong Bay. To determine missing values in the observational data, we used the Bidirectional Recurrent Imputation for Time Series (BRITS) deep learning algorithm, Auto-Regressive Integrated Moving Average (ARIMA), a widely used time series analysis method, and the Long Short-Term Memory (LSTM) deep learning method were used to predict the dissolved oxygen. We also compared accuracy of ARIMA and LSTM. The missing values were determined with high accuracy by BRITS in the surface layer; however, the accuracy was low in the lower layers. The accuracy of BRITS was unstable due to the experimental conditions in the middle layer. In the middle and bottom layers, the LSTM model showed higher accuracy than the ARIMA model, whereas the ARIMA model showed superior performance in the surface layer.

• Smart Structures and Systems
• /
• v.21 no.5
• /
• pp.705-713
• /
• 2018

High-speed rail (HSR) has been in operation and development in many countries worldwide. The explosive growth of HSR has posed great challenges for operation safety and ride comfort. Among various technological demands on high-speed trains, vibration is an inevitable problem caused by rail/wheel imperfections, vehicle dynamics, and aerodynamic instability. Ride comfort is a key factor in evaluating the operational performance of high-speed trains. In this study, online monitoring data have been acquired from an in-service high-speed train for condition assessment. The measured dynamic response signals at the floor level of a train cabin are processed by the Sperling operator, in which the ride comfort index sequence is used to identify the train's operation condition. In addition, a novel technique that incorporates salient features of Bayesian inference and time series analysis is proposed for outlier detection and change detection. The Bayesian forecasting approach enables the prediction of conditional probabilities. By integrating the Bayesian forecasting approach with time series analysis, one-step forecasting probability density functions (PDFs) can be obtained before proceeding to the next observation. The change detection is conducted by comparing the current model and the alternative model (whose mean value is shifted by a prescribed offset) to determine which one can well fit the actual observation. When the comparison results indicate that the alternative model performs better, then a potential change is detected. If the current observation is a potential outlier or change, Bayes factor and cumulative Bayes factor are derived for further identification. A significant change, if identified, implies that there is a great alteration in the train operation performance due to defects. In this study, two illustrative cases are provided to demonstrate the performance of the proposed method for condition assessment of high-speed trains.

• Journal of Korea Port Economic Association
• /
• v.32 no.2
• /
• pp.25-37
• /
• 2016

The economy of South Korea has experienced two financial crises: the 1997 Asian financial crisis and the 2008 global financial crisis. These crises had a significant impact on the nation's macro-economic indicators. Furthermore, they had a profound influence on container traffic in container ports in Busan, which is the largest port in South Korea in terms of TEUs handled. However, the impact of the Asian financial crisis on container throughput is not clear. In this study, we assume that the two financial crises are independent and different, and then analyze how each of them impacted container throughput in Busan ports. To perform this analysis, we use an intervention model that is a special type of ARIMA model with input series. Intervention models can be used to model and forecast a response series and to analyze the impact of an intervention or event on the series. This study focuses on the latter case, and our results show that the impacts of the financial crises vary considerably.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1471-1476
• /
• 2021

Flood damage can cause floods or tsunamis, which can result in enormous loss of life and property. In this regard, damage can be reduced by making a quick evacuation decision through flood prediction, and many studies are underway in this field to predict floods using time series data. In this paper, we propose a CNN-based time series prediction model. A CNN-based water level prediction model was implemented using the river level and precipitation, and the performance was confirmed by comparing it with the LSTM and GRU models, which are often used for time series prediction. In addition, by checking the performance difference according to the size of the input data, it was possible to find the points to be supplemented, and it was confirmed that better performance than LSTM and GRU could be obtained. Through this, it is thought that it can be utilized as an initial study for flood prediction.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.2
• /
• pp.348-369
• /
• 2024

With the advancement of Industry 4.0 and Industrial Internet of Things (IIoT), manufacturing increasingly seeks automation and intelligence. Temperature and vibration monitoring are essential for machinery health. Traditional abnormal state detection methodologies often overlook the intricate frequency characteristics inherent in vibration time series and are susceptible to erroneously reconstructing temperature abnormalities due to the highly similar waveforms. To address these limitations, we introduce synergistic, end-to-end, unsupervised Frequency-Time Domain Memory-Enhanced Autoencoders (FTD-MAE) capable of identifying abnormalities in both temperature and vibration datasets. This model is adept at accommodating time series with variable frequency complexities and mitigates the risk of overgeneralization. Initially, the frequency domain encoder processes the spectrogram generated through Short-Time Fourier Transform (STFT), while the time domain encoder interprets the raw time series. This results in two disparate sets of latent representations. Subsequently, these are subjected to a memory mechanism and a limiting function, which numerically constrain each memory term. These processed terms are then amalgamated to create two unified, novel representations that the decoder leverages to produce reconstructed samples. Furthermore, the model employs Spectral Entropy to dynamically assess the frequency complexity of the time series, which, in turn, calibrates the weightage attributed to the loss functions of the individual branches, thereby generating definitive abnormal scores. Through extensive experiments, FTD-MAE achieved an average ACC and F1 of 0.9826 and 0.9808 on the CMHS and CWRU datasets, respectively. Compared to the best representative model, the ACC increased by 0.2114 and the F1 by 0.1876.

• Communications for Statistical Applications and Methods
• /
• v.19 no.3
• /
• pp.507-516
• /
• 2012

We study time series models for seasonal time series data with a covariance structure that depends on time and the periodic autocorrelation at various lags $k$. In this paper, we introduce an ARMA model with periodically varying coefficients(PARMA) and analyze Arosa ozone data with a periodic correlation in the practical case study. Finally, we use a PARMA model and a seasonal ARIMA model for data analysis and show the performance of a PARMA model with a comparison to the SARIMA model.

• International Journal of Internet, Broadcasting and Communication
• /
• v.12 no.3
• /
• pp.125-130
• /
• 2020

At the end of 1997, the volatility of the exchange rate intensified as the nation's exchange rate system was converted into a free-floating exchange rate system. As a result, managing the exchange rate is becoming a very important task, and the need for forecasting the exchange rate is growing. The exchange rate prediction model using the existing exchange rate prediction method, statistical technique, cannot find a nonlinear pattern of the time series variable, and it is difficult to analyze the time series with the variability cluster phenomenon. And as the number of variables to be analyzed increases, the number of parameters to be estimated increases, and it is not easy to interpret the meaning of the estimated coefficients. Accordingly, the exchange rate prediction model using artificial neural network, rather than statistical technique, is presented. Using DNN, which is the basis of deep learning among artificial neural networks, and LSTM, a recurrent neural network model, the number of hidden layers, neurons, and activation function changes of each model found the optimal exchange rate prediction model. The study found that although there were model differences, LSTM models performed better than DNN models and performed best when the activation function was Tanh.

• International Commerce and Information Review
• /
• v.12 no.1
• /
• pp.99-121
• /
• 2010

This paper reviews the categories and properties of risk measures, analyzes the classes and structural equations of volatility forecasting models, and presents the practical methodologies and their expansion methods of estimating and forecasting the volatilities of exchange rates using Excel spreadsheet modeling. We apply the GARCH(1,1) model to the Korean won(KRW) denominated daily and monthly exchange rates of USD, JPY, EUR, GBP, CAD and CNY during the periods from January 4, 1998 to December 31, 2009, make the estimates of long-run variances in the returns of exchange rate calculated as the step-by-step change rate, and test the adequacy of estimated GARCH(1,1) model using the Box-Pierce-Ljung statistics Q and chi-square test-statistics. We demonstrate the adequacy of GARCH(1,1) model in estimating and forecasting the volatility of exchange rates in the monthly series except the semi-variance GARCH(1,1) applied to KRW/JPY100 rate. But we reject the adequacy of GARCH(1,1) model in estimating and forecasting the volatility of exchange rates in the daily series because of the very high Box-Pierce-Ljung statistics in the respective time lags resulting to the self-autocorrelation. In conclusion, the GARCH(1,1) model provides for the easy and helpful tools to forecast the exchange rate volatilities and may become the powerful methodology to overcome the application difficulties with the spreadsheet modeling.

• Water for future
• /
• v.14 no.1
• /
• pp.51-58
• /
• 1981

This study attempts to examine and estabilish a simulation model from the stochastic analysis of daily streamflow. Daily streamflow records obstained at the main gauging stations along the Han, Nakdong and Geum River were used in the analysis. The following results were abtained. From the analysis of time series of streamflow by the correlogram and spectraal density, The serial component of one-year periodicity, serial correlation and irregular or random component were found. The coefficient of determination R2 of multilag model remaine a plateau at log-two, so that second order mu.ltilag model was Known to fit in the simulation of daily streamflow, Consequently, multilag and recised Markov model of the sewnd order give the best results in simulatin of daily streamflow. But the former generally gives better results than the latter. And theoretical markev model is unfit in the simulation of daily series without modification.

• Water for future
• /
• v.18 no.1
• /
• pp.75-83
• /
• 1985

The applicability of ARMA(1, 1) multiseason model, which is in the beginning stage of active researches in the field of synthetic generation is evaluated with the streamflow data at the Nakdong stage gauging station on the main stem of the Nakdong River. The method of parameter estimation for the modelis reviewed and the statistical analysis of the generated seasonal streamflows such as corrlogram analysis and the computation of moments is made. The results obtained by ARMA(1, 1) multiseason model are compared with the historical streamflow data and also with those by two other multiseason models, namely, Thomas-Fiering model and Matalas AR(1) multiseason model. The seasonal streamflows grnerated by three multiseason models were annually summed up to form respective annual flow series whose statistics were compared with those of the annual flow series generated by three annual models, namely, AR(1), Matalas AR(1), and ARMA(1, 1) annual models. The possibility of ARMA(1, 1) multiseason model for the simultaneous generation of seasonal and annual streamflows is also evaluated.