• Journal of Intelligence and Information Systems
• /
• v.19 no.1
• /
• pp.1-17
• /
• 2013

In this article, several types of hybrid forecasting models are suggested. In particular, hybrid models using the generalized additive model (GAM) are newly suggested as an alternative to those using neural networks (NN). The prediction performances of various hybrid and non-hybrid models are evaluated using simulated time series data. Five different types of seasonal time series data related to an additive or multiplicative trend are generated over different levels of noise, and applied to the forecasting evaluation. For the simulated data with only seasonality, the autoregressive (AR) model and the hybrid AR-AR model performed equivalently very well. On the other hand, if the time series data employed a trend, the SARIMA model and some hybrid SARIMA models equivalently outperformed the others. In the comparison of GAMs and NNs, regarding the seasonal additive trend data, the SARIMA-GAM evenly performed well across the full range of noise variation, whereas the SARIMA-NN showed good performance only when the noise level was trivial.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.63 no.6
• /
• pp.131-140
• /
• 2021

It is becoming more important to utilize reclaimed lands in South Korea, due to the increasing competition for its usage among different sectors. However, the high groundwater level and poor permeability are exposing them to deterioration by salinization. Salinization is difficult to predict because the pattern changes according to various characteristics of soil and groundwater. In this study, the capillary rising time was studied by the water content profile in the soil. The prediction equation of soil salinity was developed based on simulation result of the CHEMFLO model. to enable prediction considering various soil water content and groundwater level. The two terms constituting the equation showed the coefficients of determination of 0.9816 and 0.9824, respectively. Using the prediction equation of the study, the surface salinity can be easily predicted from the initial surface salinity and the salinity of the groundwater. In the future, more precise predictions will be possible with the results of studies on the hydraulic characteristics of various reclaimed soils, changes in water content profile by seasonal and climate events.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.373-373
• /
• 2021

Rainfall forecasting is an important issue that is applied in many areas, such as agriculture, flood warning, and water resources management. In this context, this study proposed a statistical and machine learning-based forecasting model for monthly rainfall. The Bayesian Gaussian process was chosen to optimize the hyperparameters of the Stacked Long Short-term memory (SLSTM) model. The proposed SLSTM model was applied for predicting monthly precipitation of Seoul station, South Korea. Data were retrieved from the Korea Meteorological Administration (KMA) in the period between 1960 and 2019. Four schemes were examined in this study: (i) prediction with only rainfall; (ii) with deseasonalized rainfall; (iii) with rainfall and minimum temperature; (iv) with deseasonalized rainfall and minimum temperature. The error of predicted rainfall based on the root mean squared error (RMSE), 16-17 mm, is relatively small compared with the average monthly rainfall at Seoul station is 117mm. The results showed scheme (iv) gives the best prediction result. Therefore, this approach is more straightforward than the hydrological and hydraulic models, which request much more input data. The result indicated that a deep learning network could be applied successfully in the hydrology field. Overall, the proposed method is promising, given a good solution for rainfall prediction.

• Animal Bioscience
• /
• v.34 no.4
• /
• pp.770-782
• /
• 2021

Objective: The aim of the present study was to compare the effectiveness of three approaches (the seasonal auto-regressive integrated moving average [SARIMA] model, the nonlinear autoregressive exogenous [NARX] artificial neural networks and Wood's model) to the prediction of milk yield during lactation. Methods: The dataset comprised monthly test-day records from 965 Polish Holstein-Friesian Black-and-White primiparous cows. The milk yields from cows in their first lactation (from 5 to 305 days in milk) were used. Each lactation was divided into ten lactation stages of approximately 30 days. Two age groups and four calving seasons were distinguished. The records collected between 2009 and 2015 were used for model fitting and those from 2016 for the verification of predictive performance. Results: No significant differences between the predicted and the real values were found. The predictions generated by SARIMA were slightly more accurate, although they did not differ significantly from those produced by the NARX and Wood's models. SARIMA had a slightly better performance, especially in the initial periods, whereas the NARX and Wood's models in the later ones. Conclusion: The use of SARIMA was more time-consuming than that of NARX and Wood's model. The application of the SARIMA, NARX and Wood's models (after their implementation in a user-friendly software) may allow farmers to estimate milk yield of cows that begin production for the first time.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2004.05a
• /
• pp.210-215
• /
• 2004

The Purpose of this study are to develop the simulation(Tank model, Rainfall-runoff model) for the estimation of wily river discharge and for evaluation of wily pollutant loads from the watersheds of the objected basin area. As apply this constructed Tank model to Nakdong river region, we evaluated the wily river discharge of Nakdong river from use-land conditions, precipitation and evaporation data of 3 years(from 1998 to 2000) and investigate the seasonal fluctuation of SS, COD, TN, TP inflowing into Nakdong river. The result shows that summer has high pollutant level than winter in seasonal characteristic and the down stream has high pollutant level than the upper stream. The annual average of SS, COD, TN, TP flawing in Nakdong river(Samranjin) was estimated each 691ton-COD/year, 1854.2ton-SS/year, 382.8ton-TN/year and 13.0ton- TP/year.

• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.34 no.5
• /
• pp.51-58
• /
• 2018

The purpose of this study was to analyze the effect of seasonal change in characteristics of hygiene activity on domestic hot water energy consumption. With 16 residents of 4 households, the data about frequency of hygiene activity and water temperature was collected from February to August, 2017. The results of collected data discovered that the frequency of hygiene activity was higher especially in summer, whereas the consumption of warm water they used was higher in winter. The seasonal change in characteristics of hygiene activity was analyzed to be changed and strongly influenced by outdoor temperature. The influence of characteristics of hygiene activity on hot water consumption was analyzed. There was 13% of difference between consumption that was calculated taking characteristics of hygiene activity into account and consumption that was not. Therefore, this study suggested hygiene activity schedule, hot water profile and hot water consumption pattern, which can be utilized for improving simulation as well.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2000.10a
• /
• pp.159-164
• /
• 2000

In order to clarify the seasonal variation of residual current and material transportation process in Hiroshima Bay, JAPAN, the real-time simulation of residual current and particle tracking by using Euler-Lagrange model were carried out. The calculated tidal current and water temperature and salinity showed good agreement with the observed ones. The residual currents showed the southward flow pattern at the upper layer, and the northward flow pattern at the lower layer. The flow structure of residual current in Hiroshima Bay is an estuarine circulation affected by density flow and wind driven current. The residual current plays an improtant role of material transportation in th bay.

• Water for future
• /
• v.17 no.2
• /
• pp.125-131
• /
• 1984

watershed Model by mathematical formulation is one of the powerful tool to analyze the hydrologic process in a watershed. The seasonal watershed model is one of the mathematial model from which the monthly streamflow can be simulated and forcasted for given precipitaion data. This model also enables us to compute the monthly runoff at each subbgasin when the basin is subdivided into several small subbasins. The computation of runoff volume makes a Prediction of the areal distirbution of runoff volume for a given precipitation data. Several basins in Han River basin were chosen to simulate the monthly runoff and compute the runoff at each subbasin. A simple logarithmic regression were conducted between runoff ratio and area ratio. The correlation was very high and the equation can be used for prediciting flood volume when flood at downstream gaging station is know.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.35 no.5
• /
• pp.423-430
• /
• 2017

Recently, data analysis research has been carried out using the deep learning technique in various fields such as image interpretation and/or classification. Various types of algorithms are being developed for many applications. In this paper, we propose a precipitation prediction algorithm based on deep learning with high accuracy in order to take care of the possible severe damage caused by climate change. Since the geographical and seasonal characteristics of Korea are clearly distinct, the meteorological factors have repetitive patterns in a time series. Since the LSTM (Long Short-Term Memory) is a powerful algorithm for consecutive data, it was used to predict precipitation in this study. For the numerical test, we calculated the PWV (Precipitable Water Vapor) based on the tropospheric delay of the GNSS (Global Navigation Satellite System) signals, and then applied the deep learning technique to the precipitation prediction. The GNSS data was processed by scientific software with the troposphere model of Saastamoinen and the Niell mapping function. The RMSE (Root Mean Squared Error) of the precipitation prediction based on LSTM performs better than that of ANN (Artificial Neural Network). By adding GNSS-based PWV as a feature, the over-fitting that is a latent problem of deep learning was prevented considerably as discussed in this study.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.5
• /
• pp.1874-1885
• /
• 2018

The customer side operation is getting more complex in a smart grid environment because of the adoption of renewable resources. In performing energy management planning or scheduling, it is essential to forecast non-controllable resources accurately and robustly. The PV system is one of the common renewable energy resources in customer side. Its output depends on weather and physical characteristics of the PV system. Thus, weather information is essential to predict the amount of PV system output. However, weather forecast usually does not include enough solar irradiation information. In this study, a PV system power output prediction model (PPM) under limited weather information is proposed. In the proposed model, meteorological radiation model (MRM) is used to improve cloud cover radiation model (CRM) to consider the seasonal effect of the target region. The results of the proposed model are compared to the result of the conventional CRM prediction method on the PV generation obtained from a field test site. With the PPM, root mean square error (RMSE), and mean absolute error (MAE) are improved by 23.43% and 33.76%, respectively, compared to CRM for all days; while in clear days, they are improved by 53.36% and 62.90%, respectively.