• International Journal of Control, Automation, and Systems
• /
• v.6 no.5
• /
• pp.639-650
• /
• 2008

Electricity price forecasting has become an integral part of power system operation and control. In this paper, a wavelet transform (WT) based neural network (NN) model to forecast price profile in a deregulated electricity market has been presented. The historical price data has been decomposed into wavelet domain constitutive sub series using WT and then combined with the other time domain variables to form the set of input variables for the proposed forecasting model. The behavior of the wavelet domain constitutive series has been studied based on statistical analysis. It has been observed that forecasting accuracy can be improved by the use of WT in a forecasting model. Multi-scale analysis from one to seven levels of decomposition has been performed and the empirical evidence suggests that accuracy improvement is highest at third level of decomposition. Forecasting performance of the proposed model has been compared with (i) a heuristic technique, (ii) a simulation model used by Ontario's Independent Electricity System Operator (IESO), (iii) a Multiple Linear Regression (MLR) model, (iv) NN model, (v) Auto Regressive Integrated Moving Average (ARIMA) model, (vi) Dynamic Regression (DR) model, and (vii) Transfer Function (TF) model. Forecasting results show that the performance of the proposed WT based NN model is satisfactory and it can be used by the participants to respond properly as it predicts price before closing of window for submission of initial bids.

• Journal of Korea Water Resources Association
• /
• v.49 no.10
• /
• pp.877-885
• /
• 2016

In this study, the Standard Precipitation Index(SPI), meteorological drought index, was used to evaluate the temporal and spatial assessment of drought forecasting results for all cross Korea. For the drought forecasting, the Multi Layer Perceptron-Artificial Neural Network (MLP-ANN) was selected and the drought forecasting was performed according to different forecasting lead time for SPI (3) and SPI (6). The precipitation data observed in 59 gaging stations of Korea Meteorological Adminstration (KMA) from 1976~2015. For the performance evaluation of the drought forecasting, the binary classification confusion matrix, such as evaluating the status of drought occurrence based on threshold, was constituted. Then Receiver Operating Characteristics (ROC) score and F score according to conditional probability are computed. As a result of ROC analysis on forecasting performance, drought forecasting performance, of applying the MLP-ANN model, shows satisfactory forecasting results. Consequently, two-month and five-month leading forecasts were possible for SPI (3) and SPI (6), respectively.

• Journal of Korea Multimedia Society
• /
• v.22 no.9
• /
• pp.1069-1079
• /
• 2019

In this study, we developed a $PM_{10}$ forecasting model using DNN and Membership Function, and improved the forecasting performance. The model predicts the $PM_{10}$ concentrations of the next 3 days in the Seoul area by using the weather and air quality observation data and forecast data. The best model(RM14)'s accuracy (82%, 76%, 69%) and false alarm rate(FAR:14%,33%,44%) are good. Probability of detection (POD: 79%, 50%, 53%), however, are not good performance. These are due to the lack of training data for high concentration $PM_{10}$ compared to low concentration. In addition, the model dose not reflect seasonal factors closely related to the generation of high concentration $PM_{10}$. To improve this, we propose Julian date membership function as inputs of the $PM_{10}$ forecasting model. The function express a given date in 12 factors to reflect seasonal characteristics closely related to high concentration $PM_{10}$. As a result, the accuracy (79%, 70%, 66%) and FAR (24%, 48%, 46%) are slightly reduced in performance, but the POD (79%, 75%, 71%) are up to 25% improved compared with those of the RM14 model. Hence, this shows that the proposed Julian forecast model is effective for high concentration $PM_{10}$ forecasts.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.55 no.7
• /
• pp.306-312
• /
• 2006

Support Vector Machine(SVM), of which the foundations have been developed by Vapnik (1995), is gaining popularity thanks to many attractive features and promising empirical performance. In this paper, we propose a new short-term load forecasting technique based on SVM. We discuss the input vector selection of SVM for load forecasting and analyze the prediction performance for various SVM parameters such as kernel function, cost coefficient C, and $\varepsilon$ (the width of 8 $\varepsilon-tube$). The computer simulation shows that the prediction performance of the proposed method is superior to that of the conventional neural networks.

• Journal of Environmental Science International
• /
• v.24 no.8
• /
• pp.1023-1036
• /
• 2015

A reliable streamflow forecasting is essential for flood disaster prevention, reservoir operation, water supply and water resources management. This study proposes a hybrid model for river stage forecasting and investigates its accuracy. The proposed model is the wavelet packet-based artificial neural network(WPANN). Wavelet packet transform(WPT) module in WPANN model is employed to decompose an input time series into approximation and detail components. The decomposed time series are then used as inputs of artificial neural network(ANN) module in WPANN model. Based on model performance indexes, WPANN models are found to produce better efficiency than ANN model. WPANN-sym10 model yields the best performance among all other models. It is found that WPT improves the accuracy of ANN model. The results obtained from this study indicate that the conjunction of WPT and ANN can improve the efficiency of ANN model and can be a potential tool for forecasting river stage more accurately.

• Journal of Intelligence and Information Systems
• /
• v.1 no.1
• /
• pp.135-148
• /
• 1995

Neural Networks have been advocated as an alternative to statistical forecasting methods. However, the empirical evidences are not consistent. In the present experiments, multi-layered perceptron (MLP) are adopted as approximator to the time series generating processes. To prevent the MLP from being overfitted to the given time series, the information obtained from ARMA modeling is used to determine the architecture of MLP. The proposed approach was tested empirically using the subsamples of the 111 time series used in the first Markridakis Competition. The forecasting results were analyzed to find out the factors that affect the performance of MLP. The experimental results show that the proposed approach outperforms ARMA models in terms of fitting and forecasting accuracy. In addition, it is found that the use of deseasonalized data improves the forecasting accuracy of MLP.

• Journal of Korean Society of Transportation
• /
• v.33 no.1
• /
• pp.61-69
• /
• 2015

This study compares the forecasting accuracy of five econometric models on domestic total freight volume in South Korea. Applied five models are as follows: Ordinary Least Square model, Partial Adjustment model, Reduced Autoregressive Distributed Lag model, Vector Autoregressive model, Time Varying Parameter model. Estimating models and forecasting are carried out based on annual data of domestic freight volume and an index of industrial production during 1970~2011. 1-year, 3-year, and 5-year ahead forecasting performance of five models was compared using the recursive forecasting method. Additionally, two forecasting periods were set to compare forecasting accuracy according to the size of future volatility. As a result, the Time Varying Parameter model showed the best accuracy for forecasting periods having fluctuations, whereas the Vector Autoregressive model showed better performance for forecasting periods with gradual changes.

• The Korean Journal of Financial Management
• /
• v.20 no.1
• /
• pp.93-124
• /
• 2003

This paper studies the financial analyst's forecasting activities on the firm's operating performance during the period from 1999 to 2003. In this study, financial analyst's forecasting activities are focused on the sales, operating income and net income and financial analyst's forecasting accuracy, forecasting revising patterns and forecasting activities to the unexpected firm's operating performance are studied. Some empirical findings in this study are as follows. First, standard estimate error on the sales, operating income and net income are all significantly negative value and so financial analyst's forecast on the firm's operating performance are upwardly biased. Second, domestic financial analyst's forecasting activities is relatively more accuracy than foreign financial analyst's forecasting activities. Third, forecasting time is more close to the end of the operating performance announcement day, forecasting activities are more accuracy. Fourth, comparing with individual financial analyst's forecast, consensus forecast is more accuracy. Fifth, in the comparative forecasting activities study according to the prior firm's operating performance, financial analyst's forecasting revision activities are found to be upward or downward. Sixth, financial analysts overreact in the sales forecast and underreact in the operating income and net income forecast. Seventh, in the empirical analysis on the Easterwood-Nutt's test model(1999) which the firm's performance change are divided into the expected performance change and the unexpected performance change, it is found that financial analyst's forecasting activities on the firm's operating performance are systematically optimistic.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.4_2
• /
• pp.715-724
• /
• 2023

Solar energy forecasting is essential for (1) power system planning, management, and operation, requiring accurate predictions. It is crucial for (2) ensuring a continuous and sustainable power supply to customers and (3) optimizing the operation and control of renewable energy systems and the electricity market. Recently, research has been focusing on developing solar energy forecasting models that can provide daily plans for power usage and production and be verified in the electricity market. In these prediction models, various data, including solar energy generation and climate data, are chosen to be utilized in the forecasting process. The most commonly used climate data (such as temperature, relative humidity, precipitation, solar radiation, and wind speed) significantly influence the fluctuations in solar energy generation based on weather conditions. Therefore, this paper proposes a hybrid forecasting model by combining the strengths of the Prophet model and the GRU model, which exhibits excellent predictive performance. The forecasting periods for solar energy generation are tested in short-term (2 days, 7 days) and medium-term (15 days, 30 days) scenarios. The experimental results demonstrate that the proposed approach outperforms the conventional Prophet model by more than twice in terms of Root Mean Square Error (RMSE) and surpasses the modified GRU model by more than 1.5 times, showcasing superior performance.

• Journal of Technology Innovation
• /
• v.23 no.2
• /
• pp.33-54
• /
• 2015

Supercomputing performance demand forecasting at the national level is an important information to the researchers in fields of the computational science field, the specialized agencies which establish and operate R&D infrastructure, and the government agencies which establish science and technology infrastructure. This study derived the factors affecting the scientific and technological capability through the analysis of supercomputing performance prediction research, and it proposed a hybrid forecasting model of applying the super-computer technology trends. In the cross-sectional analysis, multiple regression analysis was performed using factors with GDP, GERD, the number of researchers, and the number of SCI papers that could affect the supercomputing performance. In addition, the supercomputing performance was predicted by multiplying in the cross-section analysis with technical progress rate of time period which was calculated by time series analysis using performance(Rmax) of Top500 data. Korea's performance scale of supercomputing in 2016 was predicted using the proposed forecasting model based on data of the top500 supercomputer and supercomputing performance demand in Korea was predicted using a cross-sectional analysis and technical progress rate. The results of this study showed that the supercomputing performance is expected to require 15~30PF when it uses the current trend, and is expected to require 20~40PF when it uses the trend of the targeting national-level. These two results showed significant differences between the forecasting value(9.6PF) of regression analysis and the forecasting value(2.5PF) of cross-sectional analysis.