• Atmosphere
• /
• v.31 no.1
• /
• pp.73-83
• /
• 2021

Numerical weather prediction (NWP) models play an essential role in predicting weather factors, but using them is challenging due to various factors. To overcome the difficulties of NWP models, deep learning models have been deployed in weather forecasting by several recent studies. This study adapts long short-term memory (LSTM), which demonstrates remarkable performance in time-series prediction. The combination of LSTM model input of meteorological features and activation functions have a significant impact on the performance therefore, the results from 5 combinations of input features and 4 activation functions are analyzed in 9 Automated Surface Observing System (ASOS) stations corresponding to cities/islands/mountains. The optimized LSTM model produces better performance within eight forecast hours than Local Data Assimilation and Prediction System (LDAPS) operated by Korean meteorological administration. Therefore, this study illustrates that this LSTM model can be usefully applied to very short-term weather forecasting, and further studies about CNN-LSTM model with 2-D spatial convolution neural network (CNN) coupled in LSTM are required for improvement.

• Wind and Structures
• /
• v.36 no.6
• /
• pp.379-392
• /
• 2023

Structural health monitoring is used to ensure the well-being of civil structures by detecting damage and estimating deterioration. Wind flow applies external loads to high-rise buildings, with the horizontal force component of the wind causing structural displacements in high-rise buildings. This study proposes a deep learning-based predictive model for measuring lateral displacement response in high-rise buildings. The proposed long short-term memory model functions as a sequence generator to generate displacements on building floors depending on the displacement statistics collected on the top floor. The model was trained with wind-induced displacement data for the top floor of a high-rise building as input. The outcomes demonstrate that the model can forecast wind-induced displacement on the remaining floors of a building. Further, displacement was predicted for each floor of the high-rise buildings at wind flow angles of 0° and 45°. The proposed model accurately predicted a high-rise building model's story drift and lateral displacement. The outcomes of this proposed work are anticipated to serve as a guide for assessing the overall lateral displacement of high-rise buildings.

• International Journal of Advanced Culture Technology
• /
• v.11 no.4
• /
• pp.393-405
• /
• 2023

Accurate prediction of wind speed and power is vital for enhancing the efficiency of wind energy systems. Numerous solutions have been implemented to date, demonstrating their potential to improve forecasting. Among these, deep learning is perceived as a revolutionary approach in the field. However, despite their effectiveness, the noise present in the collected data remains a significant challenge. This noise has the potential to diminish the performance of these algorithms, leading to inaccurate predictions. In response to this, this study explores a novel feature engineering approach. This approach involves altering the data input shape in both Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) and Autoregressive models for various forecasting horizons. The results reveal substantial enhancements in model resilience against noise resulting from step increases in data. The approach could achieve an impressive 83% accuracy in predicting unseen data up to the 24th steps. Furthermore, this method consistently provides high accuracy for short, mid, and long-term forecasts, outperforming the performance of individual models. These findings pave the way for further research on noise reduction strategies at different forecasting horizons through shape-wise feature engineering.

• Wind and Structures
• /
• v.38 no.1
• /
• pp.75-91
• /
• 2024

Reliable wind signal reconstruction can be beneficial to the operational safety of long-span bridges. Non-Gaussian characteristics of wind signals make the reconstruction process challenging. In this paper, non-Gaussian wind signals are converted into a combined prediction of two kinds of features, actual wind speeds and wind angles of attack. First, two decomposition techniques, empirical mode decomposition (EMD) and variational mode decomposition (VMD), are introduced to decompose wind signals into intrinsic mode functions (IMFs) to reduce the randomness of wind signals. Their principles and applicability are also discussed. Then, four artificial intelligence (AI) algorithms are utilized for wind signal reconstruction by combining the particle swarm optimization (PSO) algorithm with back propagation neural network (BPNN), support vector regression (SVR), long short-term memory (LSTM) and bidirectional long short-term memory (Bi-LSTM), respectively. Measured wind signals from a bridge site in a deep-cutting gorge are taken as experimental subjects. The results showed that the reconstruction error of high-frequency components of EMD is too large. On the contrary, VMD fully extracts the multiscale rules of the signal, reduces the component complexity. The combination of VMD-PSO-Bi-LSTM is demonstrated to be the most effective among all hybrid models.

• Geomechanics and Engineering
• /
• v.37 no.4
• /
• pp.395-418
• /
• 2024

The epicentral region of earthquakes is typically where liquefaction-related damage takes place. To determine the maximum distance, such as maximum epicentral distance (R_e), maximum fault distance (R_f), or maximum hypocentral distance (R_h), at which an earthquake can inflict damage, given its magnitude, this study, using a recently updated global liquefaction database, multiple ML models are built to predict the limiting distances (R_e, R_f, or R_h) required for an earthquake of a given magnitude to cause damage. Four machine learning models LSTM (Long Short-Term Memory), BiLSTM (Bidirectional Long Short-Term Memory), CNN (Convolutional Neural Network), and XGB (Extreme Gradient Boosting) are developed using the Python programming language. All four proposed ML models performed better than empirical models for limiting distance assessment. Among these models, the XGB model outperformed all the models. In order to determine how well the suggested models can predict limiting distances, a number of statistical parameters have been studied. To compare the accuracy of the proposed models, rank analysis, error matrix, and Taylor diagram have been developed. The ML models proposed in this paper are more robust than other current models and may be used to assess the minimal energy of a liquefaction disaster caused by an earthquake or to estimate the maximum distance of a liquefied site provided an earthquake in rapid disaster mapping.

• Journal of Bio-Environment Control
• /
• v.32 no.4
• /
• pp.384-395
• /
• 2023

Since transpiration plays a key role in optimal irrigation management, knowledge of the irrigation demand of crops like tomatoes, which are highly susceptible to water stress, is necessary. One way to determine irrigation demand is to measure transpiration, which is affected by environmental factor or growth stage. This study aimed to estimate the transpiration amount of tomatoes and find a suitable model using mathematical and deep learning models using minute-by-minute data. Pearson correlation revealed that observed environmental variables significantly correlate with crop transpiration. Inside air temperature and outside radiation positively correlated with transpiration, while humidity showed a negative correlation. Multiple Linear Regression (MLR), Polynomial Regression model, Artificial Neural Network (ANN), Long short-term Memory (LSTM), and Gated Recurrent Unit (GRU) models were built and compared their accuracies. All models showed potential in estimating transpiration with R² values ranging from 0.770 to 0.948 and RMSE of 0.495 mm/min to 1.038 mm/min in the test dataset. Deep learning models outperformed the mathematical models; the GRU demonstrated the best performance in the test data with 0.948 R² and 0.495 mm/min RMSE. The LSTM and ANN closely followed with R² values of 0.946 and 0.944, respectively, and RMSE of 0.504 m/min and 0.511, respectively. The GRU model exhibited superior performance in short-term forecasts while LSTM for long-term but requires verification using a large dataset. Compared to the FAO56 Penman-Monteith (PM) equation, PM has a lower RMSE of 0.598 mm/min than MLR and Polynomial models degrees 2 and 3 but performed least among all models in capturing variability in transpiration. Therefore, this study recommended GRU and LSTM models for short-term estimation of tomato transpiration in greenhouses.

• Journal of the Korean Solar Energy Society
• /
• v.26 no.2
• /
• pp.27-34
• /
• 2006

In this paper, a forecasting model of wind speed at Walryong Site, Jeju Island is presented, which has been developed and evaluated as a first step toward establishing Korea Forecasting Model of Wind Power Generation. The forecasting model is constructed based on neural network and is trained with wind speed data observed at Cosan Weather Station located near by Walryong Site. Due to short period of measurements at Walryong Site for training statistical model Gosan Weather Station's long-term data are substituted and then transplanted to Walryong Site by using Measure-Correlate-Predict technique. One to three-hour advance forecasting of wind speed show good agreements with the monitoring data of Walryong site with the correlation factors 0.96 and 0.88, respectively.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.54 no.3
• /
• pp.144-151
• /
• 2005

This paper presents a forecasting technique of the short-term marginal price (SMP) using an Artificial Neural Network (ANN). The SW forecasting is a very important element in an electricity market for the optimal biddings of market participants as well as for market stabilization of regulatory bodies. Input data are organized in two different approaches, time-axis and day-axis approaches, and the resulting patterns are used to train the ANN. Performances of the two approaches are compared and the better estimate is selected by a composition rule to forecast the SMP. By combining the two approaches, the proposed composition technique reflects the characteristics of hourly, daily and seasonal variations, as well as the condition of sudden changes in the spot market, and thus improves the accuracy of forecasting. The proposed method is applied to the historical real-world data from the Korea Power Exchange (KPX) to verify the effectiveness of the technique.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.53 no.10
• /
• pp.529-535
• /
• 2004

Load forecasting is essential in the electricity market for the participants to manage the market efficiently and stably. A wide variety of techniques/algorithms for load forecasting has been reported in many literatures. These techniques are as follows: multiple linear regression, stochastic time series, general exponential smoothing, state space and Kalman filter, knowledge-based expert system approach (fuzzy method and artificial neural network). These techniques have improved the accuracy of the load forecasting. In recent 10 years, many researchers have focused on artificial neural network and fuzzy method for the load forecasting. In this paper, we propose an algorithm of a hybrid load forecasting method using fuzzy linear regression and general exponential smoothing and considering the sensitivities of the temperature. In order to consider the lower load of weekends and Monday than weekdays, fuzzy linear regression method is proposed. The temperature sensitivity is used to improve the accuracy of the load forecasting through the relation of the daily load and temperature. And the normal load of weekdays is easily forecasted by general exponential smoothing method. Test results show that the proposed algorithm improves the accuracy of the load forecasting in 1996.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.7
• /
• pp.2785-2799
• /
• 2020

We propose a novel speaker adaptation method using acoustic model clustering. The similarity of different speakers is defined by the cosine distance between their i-vectors (intermediate vectors), and various efficient clustering algorithms are applied to obtain a number of speaker subsets with different characteristics. The speaker-independent model is then retrained with the training data of the individual speaker subsets grouped by the clustering results, and an unknown speech is recognized by the retrained model of the closest cluster. The proposed method is applied to a large-scale speech recognition system implemented by a hybrid hidden Markov model and deep neural network framework. An experiment was conducted to evaluate the word error rates using Resource Management database. When the proposed speaker adaptation method using i-vector based clustering was applied, the performance, as compared to that of the conventional speaker-independent speech recognition model, was improved relatively by as much as 12.2% for the conventional fully neural network, and by as much as 10.5% for the bidirectional long short-term memory.