• Journal of KIISE
• /
• v.44 no.9
• /
• pp.954-965
• /
• 2017

A stable power supply is very important for the maintenance and operation of the power infrastructure. Accurate power consumption prediction is therefore needed. In particular, a university campus is an institution with one of the highest power consumptions and tends to have a wide variation of electrical load depending on time and environment. For this reason, a model that can accurately predict power consumption is required for the effective operation of the power system. The disadvantage of the existing time series prediction technique is that the prediction performance is greatly degraded because the width of the prediction interval increases as the difference between the learning time and the prediction time increases. In this paper, we first classify power data with similar time series patterns considering the date, day of the week, holiday, and semester. Next, each ARIMA model is constructed based on the classified data set and a daily power consumption forecasting method of the university campus is proposed through the time series cross-validation of the predicted time. In order to evaluate the accuracy of the prediction, we confirmed the validity of the proposed method by applying performance indicators.

• Journal of Intelligence and Information Systems
• /
• v.4 no.2
• /
• pp.23-34
• /
• 1998

Travel time forecasting, especially public bus travel time forecasting in urban areas, is a difficult and complex problem which requires a prohibitively large computation time and years of experience. As the network of target area grows with addition of streets and lanes, computational burden of the forecasting systems exponentially increases. Even though the travel time between two neighboring intersections is known a priori, it is still difficult, if not impossible, to compute the travel time between every two intersections. For the reason, previous approaches frequently have oversimplified the transportation network to show feasibilities of the problem solving algorithms. In this paper, forecasting of the travel time between every two intersections is attempted based on travel time data between two neighboring intersections. The time stamps data of public buses which recorded arrival time at predetermined bus stops was extensively collected and forecast. At first, the time stamp data was categorized to eliminate white noise, uncontrollable in forecasting, based on wavelet conversion. Then, the radial basis neural networks was applied to remaining data, which showed relatively accurate results. The success of the attempt was confirmed by the drastically reduced relative error when the nodes between the target intersections increases. In general, as the number of the nodes between target intersections increases, the relative error shows the tendency of sharp increase. The experimental results of the novel approaches, based on wavelet conversion and neural network teaming mechanism, showed the forecasting methodology is very promising.

• The Korean Journal of Applied Statistics
• /
• v.29 no.1
• /
• pp.193-203
• /
• 2016

This paper forecasts electricity demand as a critical element of a demand management system in Smart Grid environment. We present a prediction method of using a combination of predictive values by time series clustering. Periodogram-based normalized clustering, predictive analysis clustering and dynamic time warping (DTW) clustering are proposed for time series clustering methods. Double Seasonal Holt-Winters (DSHW), Trigonometric, Box-Cox transform, ARMA errors, Trend and Seasonal components (TBATS), Fractional ARIMA (FARIMA) are used for demand forecasting based on clustering. Results show that the time series clustering method provides a better performances than the method using total amount of electricity demand in terms of the Mean Absolute Percentage Error (MAPE).

• Proceedings of the KIEE Conference
• /
• 1995.07b
• /
• pp.578-580
• /
• 1995

In this paper, a method for the daily maximum load forecasting which uses a chaotic time series in power system and artificial neural network. We find the characteristics of chaos in power load curve and then determine a optimal embedding dimension and delay time, For the load forecast of one day ahead daily maximum power, we use the time series load data obtained in previous year. By using of embedding dimension and delay time, we construct a strange attractor in pseudo phase plane and the artificial neural network model trained with the attractor font mentioned above. The one day ahead forecast errors are about 1.4% of absolute percentage average error.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.835-837
• /
• 1996

In this paper, a method for the daily maximum load forecasting which uses a chaotic time series in power system and artificial neural network(Back-propagation) is proposed. We find the characteristics of chaos in power load curve and then determine a optimal embedding dimension and delay time. For the load forecast of one day ahead daily maximum power, we use the time series load data obtained in previous year. By using of embedding dimension and delay time, we construct a strange attractor in pseudo phase plane and the artificial neural network model trained with the attractor mentioned above. The one day ahead forecast errors are about 1.4% for absolute percentage average error.

• The Journal of Bigdata
• /
• v.9 no.1
• /
• pp.31-47
• /
• 2024

With the advancement of artificial intelligence, the travel and hospitality industry is also adopting AI and machine learning technologies for various purposes. In the tourism industry, demand forecasting is recognized as a very important factor, as it directly impacts service efficiency and revenue maximization. Demand forecasting requires the consideration of time-varying data flows, which is why statistical techniques and machine learning models are used. In recent years, variations and integration of existing models have been studied to account for the diversity of demand forecasting data and the complexity of the natural world, which have been reported to improve forecasting performance concerning uncertainty and variability. This study also proposes a new model that integrates various machine-learning approaches to improve the accuracy of hotel sales demand forecasting. Specifically, this study proposes a new time series forecasting model based on XGBoost that selectively utilizes a local model by clustering with DTW K-means and a global model using the entire data to improve forecasting performance. The hotel demand forecasting model that selectively utilizes global and regional models proposed in this study is expected to impact the growth of the hotel and travel industry positively and can be applied to forecasting in other business fields in the future.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.12
• /
• pp.579-586
• /
• 2021

Time series forecasting refers to predicting future time information based on past time information. Accurately predicting future information is crucial because it is used for establishing strategies or making policy decisions in various fields. Recently, a transformer model has been mainly studied for a time series prediction model. However, the existing transformer model has a limitation in that it has an auto-regressive structure in which the output result is input again when the prediction sequence is output. This limitation causes a problem in that accuracy is lowered when predicting a distant time point. This paper proposes a sequential decoding model focusing on the style transformation technique to handle these problems and make more precise time series forecasting. The proposed model has a structure in which the contents of past data are extracted from the transformer-encoder and reflected in the style-based decoder to generate the predictive sequence. Unlike the decoder structure of the conventional auto-regressive transformer, this structure has the advantage of being able to more accurately predict information from a distant view because the prediction sequence is output all at once. As a result of conducting a prediction experiment with various time series datasets with different data characteristics, it was shown that the model presented in this paper has better prediction accuracy than other existing time series prediction models.

• International Journal of Contents
• /
• v.17 no.3
• /
• pp.1-14
• /
• 2021

Since the outbreak of the coronavirus disease 2019 (COVID-19) pandemic, a lot of efforts have been made in the field of data science to help combat against this disease. Among them, forecasting the number of cases of infection is a crucial problem to predict the development of the pandemic. Many deep learning-based models can be applied to solve this type of time series problem. In this research, we would like to take a step forward to incorporate spatial data (geography) with time series data to forecast the cases of region-level infection simultaneously. Specifically, we model a single spatio-temporal graph, in which nodes represent the geographic regions, spatial edges represent the distance between each pair of regions, and temporal edges indicate the node features through time. We evaluate this approach in COVID-19 in a Korean dataset, and we show a decrease of approximately 10% in both RMSE and MAE, and a significant boost to the training speed compared to the baseline models. Moreover, the training efficiency allows this approach to be extended for a large-scale spatio-temporal dataset.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.2
• /
• pp.321-332
• /
• 2013

Because of the continuous low birthrates, the number of the elementary students will decrease by 17% in 2020 compared to 2011. The purpose of this study is to forecast the number of elementary school teachers until 2020. We used the data in education statistical year books from 1970 to 2010. We used the time-series regression model, time series grouped regression model and exponential smoothing model to predict the number of teachers for the next ten years. Consequently time-series grouped regression model is a better model for forecasting the number of elementary school teachers than other models.

• East Asian Journal of Business Economics (EAJBE)
• /
• v.3 no.3
• /
• pp.14-26
• /
• 2015

A wind speed forecast is a crucial and sophisticated task in a wind farm for planning turbines and corresponds to an estimate of the expected production of one or more wind turbines in the near future. By production is often meant available power for wind farm considered (with units KW or MW depending on both the wind speed and direction. Such forecasts can also be expressed in terms of energy, by integrating power production over each time interval. In this study, we technically focused on mathematical modeling of wind speed and direction forecast based on locally data set gathered from Aghdasiyeh station in Tehran. The methodology is set on using most common techniques derived from literature review. Hence we applied the most sophisticated forecasting methods to embed seasonality, trend, and irregular pattern for wind speed as an angular variables. Through this research, we carried out the most common techniques such as the Box and Jenkins family, VARMA, the component method, the Weibull function and the Fourier series. Finally, the best fit for each forecasting method validated statistically based on white noise properties and the final comparisons using residual standard errors and mean absolute deviation from real data.