• 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.

• The Korean Journal of Applied Statistics
• /
• v.33 no.2
• /
• pp.147-160
• /
• 2020

In this paper, we study an algorithm that automatically determines the orders of past observations and conditional mean values that play an important role in count time series models. Based on the orders of the ARIMA model, the algorithm constitutes the order candidates group for time series generalized linear models and selects the final model based on information criterion among the combinations of the order candidates group. To evaluate the proposed algorithm, we perform small simulations and empirical analysis according to underlying models and time series as well as compare forecasting performances with the ARIMA model. The results of the comparison confirm that the time series generalized linear model offers better performance than the ARIMA model for the count time series analysis. In addition, the empirical analysis shows better performance in mid and long term forecasting than the ARIMA model.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.12
• /
• pp.8576-8584
• /
• 2015

Load forecasting is needed to make supply and demand plan for a stable supply of electricity. It is also necessary for optimal operational plan of the power system planning. In particular, in order to ensure stable power supply, long-term load forecasting is important. And regional load forecasting is important for tightening supply stability. Regional load forecasting is known to be an essential process for the optimal state composition and maintenance of the electric power system network including transmission lines and substations to meet the load required for the area. Therefore, in this paper we propose a forecasting method using SARIMA during the 12 months (long-term/mid-term) load forecasting by 16 regions of the South Korea.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.8
• /
• pp.1647-1652
• /
• 2009

Generally the neural network and the fuzzy compensative algorithm are applied to forecast the time series for power demand with a characteristic of non-linear dynamic system, but it has a few prediction errors relatively. It also makes long term forecast difficult for sensitivity on the initial condition. On this paper, we evaluate the chaotic characteristic of electrical power demand with analysis methods of qualitative and quantitative and perform a forecast simulation of electrical power demand in regular sequence, attractor reconstruction, time series forecast for multi dimension using Lyapunov exponent quantitatively. We compare simulated results with the previous method and verify that the purpose one being more practice and effective than it.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.05a
• /
• pp.171-174
• /
• 2009

Generally the neural network and the fuzzy compensative algorithm are applied to forecast the time series for power demand with a characteristic of non-linear dynamic system, but it has a few prediction errors relatively. It also makes long term forecast difficult for sensitivity on the initial condition. On this paper, we evaluate the chaotic characteristic of electrical power demand with analysis methods of qualitative and quantitative and perform a forecast simulation of electrical power demand in regular sequence, attractor reconstruction, time series forecast for multi dimension using Lyapunov exponent quantitatively. We compare simulated results with the previous method and verify that the purpose one being more practice and effective than it.

• Journal of the Society of Naval Architects of Korea
• /
• v.40 no.5
• /
• pp.60-67
• /
• 2003

For risk or reliability assessment of ship structures against particular hazardous situations such as total loss or sinking due to hull girder collapse, the short-term based response analysis rather than the long-term response analysis is required to determine wave-induced bending moments when the ship encounters a storm of specific duration and with a specified small encounter probability. In the present study, the effects of operational condition and sea states on wave-induced bending moments of large merchant vessels are investigated. A series of the short-term response analyses for a hypothetical VLCC and a Capesize bulk carrier (CSBC) are carried out with varying operational condition and sea states which include ship speed, significant wave height and wave persistence time, using the linear-strip theory based program ABS/SHIPMOTION and the MIT sea-keeping tables. The computed results are also compared with the IACS design formula predictions. The results and insights developed from the present study are summarized.

• The Journal of Asian Finance, Economics and Business
• /
• v.9 no.4
• /
• pp.239-250
• /
• 2022

This study examined the level of capital structure and its determinants of publicly traded companies in Malaysia before and after the COVID-19 pandemic. The data for this study was examined using Python Programming Language and time-series financial data from 2,784 quarterly observations in 2019 and 2020. The maximum debt is larger before the COVID-19 period, according to the findings. During the COVID-19 period, short-term debts and total debts have both decreased slightly. However, long-term debts have increased marginally. As a result, this research demonstrates that the capital structure has changed slightly during the COVID-19 period. The findings imply that independent of the capital structure proxies, tangibility, liquidity, and business size had an impact on capital structure in both periods. It was found that profitability had a significant impact on total debts both before and after the COVID-19 crisis. While higher-profit enterprises appear to have lesser short-term debts before the COVID-19 periods, they are also more likely to have lower long-term debts during the COVID-19 periods. Even though growing companies tend to have higher short-term debts and thus total debts during those periods, longterm debts are unaffected by potential growth.

• Journal of Korea Water Resources Association
• /
• v.44 no.12
• /
• pp.967-974
• /
• 2011

Based on daily time series from RDAPS numerical weather forecast, Streamflow prediction was simulated and the result of ESP analysis was implemented considering quantitative mid- and long-term forecast to compare the results and review applicability. The result of ESP, ESP considering quantitative weather forecast, and flow forecast from RDAPS numerical weather forecast were compared and analyzed with average observed streamflow in Guem River Basin. Through this process, the improvement effect per method was estimated. The result of ESP considering weather information was satisfactory relatively based on long-term flow forecast simulation result. Discrepancy ratio analysis for estimating accuracy of probability forecast had similar result. It is expected to simulate more accurate flow forecast for RDAPS numerical weather forecast with improved daily scenario including time resolution, which is able to accumulate 3 hours rainfall or continuous simulation estimation.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.34 no.4
• /
• pp.457-465
• /
• 2010

In this study, a Taylor series (T-S) model based on the Arrhenius, McVetty, and Monkman-Grant equations was developed using a mathematical analysis. In order to reduce fitting errors, the McVetty equation was transformed by considering the first three terms of the Taylor series equation. The model parameters were accurately determined by a statistical technique of maximum likelihood estimation, and this model was applied to the creep data of alloy 617. The T-S model results showed better agreement with the experimental data than other models such as the Eno, exponential, and L-M models. In particular, the T-S model was converted into an isothermal Taylor series (IT-S) model that can predict the creep strength at a given temperature. It was identified that the estimations obtained using the converted ITS model was better than that obtained using the T-S model for predicting the long-term creep life of alloy 617.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.5
• /
• pp.1-18
• /
• 2023

With the recent developments in big data and deep learning, a variety of traffic information is collected widely and used for traffic operations. In particular, long short-term memory (LSTM) is used in the field of traffic information prediction with time series characteristics. Since trends, seasons, and cycles differ due to the nature of time series data input for an LSTM, a trial-and-error method based on characteristics of the data is essential for prediction models based on time series data in order to find hyperparameters. If a methodology is established to find suitable hyperparameters, it is possible to reduce the time spent in constructing high-accuracy models. Therefore, in this study, a traffic information prediction model is developed based on highway vehicle detection system (VDS) data and LSTM, and an impact assessment is conducted through changes in the LSTM evaluation indicators for each hyperparameter. In addition, a methodology for finding hyperparameters suitable for predicting highway traffic information in the transportation field is presented.