• 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.

• Journal of The Korean Association For Science Education
• /
• v.29 no.3
• /
• pp.348-358
• /
• 2009

The purpose of this study was to investigate brain activation pattern and functional connectivity network during experimental design on the biological phenomena. Twenty six right-handed healthy science teachers volunteered to be in the present study. To investigate participants' brain activities during the tasks, 3.0T fMRI system with the block experimental-design was used to measure BOLD signals of their brain and SPM2 software package was applied to analyze the acquired initial image data from the fMRI system. According to the analyzed data, superior, middle and inferior frontal gyrus, superior and inferior parietal lobule, fusiform gyrus, lingual gyrus, and bilateral cerebellum were significantly activated during participants' carrying-out experimental design. The network model was consisting of six nodes (ROIs) and its six connections. These results suggested the notion that the activation and connections of these regions mean that experimental design process couldn't succeed just a memory retrieval process. These results enable the scientific experimental design process to be examined from the cognitive neuroscience perspective, and may be used as a basis for developing a teaching-learning program for scientific experimental design such as brain-based science education curriculum.

• Structural Engineering and Mechanics
• /
• v.92 no.2
• /
• pp.121-131
• /
• 2024

This paper proposes a fault detection method for blade pitch systems of floating wind turbines using transformer-based deep-learning models. Transformers leverage self-attention mechanisms, efficiently process time-series data, and capture long-term dependencies more effectively than traditional recurrent neural networks (RNNs). The model was trained using normal operational data to detect anomalies through high reconstruction losses when encountering abnormal data. In this study, various fault conditions in a blade pitch system, including environmental load cases, were simulated using a detailed model of a spar-type floating wind turbine, the data collected from these simulations were used to train and test the transformer models. The model demonstrated superior fault-detection capabilities with high accuracy, precision, recall, and F1 scores. The results show that the proposed method successfully identifies faults and achieves high-performance metrics, outperforming existing traditional multi-layer perceptron (MLP) models and long short-term memory-autoencoder (LSTM-AE) models. This study highlights the potential of transformer models for real-time fault detection in wind turbines, contributing to more advanced condition-monitoring systems with minimal human intervention.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.47 no.5
• /
• pp.14-27
• /
• 2019

This study presents a plan for a memorial park that respects the characteristics based on the historical facts for the concept of space of the Gwangju Hospital of Korea Army and the location of the 505 Security Forces, which were designated as historic sites after the 5-18 Democratization Movement. The Gwangju Metropolitan City as it is the location of the 5-18 historic sites, is taking part in the 5-18 Memorial Project, and plans to establish a city park recognizing the historic site of the 5-18 Democratization Movement, which has been preserved only as a memory space to this point. The park is promoting a phased development plan. This study suggests that the 5-18 historic sites can be modernized and that social consensus can establish the framework of the step-by-step planning and composition process to ensure the plans for the space heals wounds while preserving the history. In this paper, we propose a solution to a problem. We solve the approach for space utilization through an analysis of precedent research and planning cases related to park planning at historical sites. In addition to exploring the value of the site, we also describe the space utilization strategy that covers the historical characteristics and facts while maintaining the concept of park planning. As a result of the research, the historic site of the Gwangju Hospital of Korea Army is planned as a park of historical memory and healing in order to solve the problems left behind by the 5-18 Democratization Movement. The historic site of the 505 Security Forces was selected as an area for historical experiences and a place for learning that can be sympathized with by future generations of children and adolescents in terms of expanding and sustaining the memory of the 5-18 Democratization Movement. In the planning stage, the historical sites suggested the direction of space utilization for representation as did the social consensus of citizens, related groups, and specialists. Through this study, we will contribute to construction of a memorial park containing historical values in from 5-18 historic sites. It is meaningful to suggest a direction that can revitalize the life of the city as well as its citizen and can share with the history with future generations beyond being a place to heal wounds and keep alive the memory of the past.

• English Language & Literature Teaching
• /
• v.17 no.3
• /
• pp.27-56
• /
• 2011

The analysis on the use of lexical discovery and consolidation strategies that have been researched within the area of vocabulary learning strategies (VLS) have not sufficiently drawn the interest of EFL practitioners with regard to vocational high school learners. The results, however, are expected to have implications for the design of vocabulary tasks and instructional materials for EFL learners. The present study investigates EFL vocational high school learners' use of lexical discovery and consolidation strategies with questionnaires, where the use of the learners' lexical discovery strategies were further validated with the think-aloud methodology by asking samples of proficient and less proficient learners to report on their reading process while reading L2 texts that had not been exposed to the learners. The results indicated that there were significant differences between the two groups of learners in the employment of 11 of the strategies which were in the categories of determination, social, memory, and metacognitive strategies, but not for cognitive strategies. The pattern of strategies indicated that different lexical discovery and consolidation strategies were employed relatively more by one proficiency group than another. The study suggests some implications for how strategy-based instruction can be implemented in EFL classrooms.

• Biomolecules & Therapeutics
• /
• v.18 no.1
• /
• pp.24-31
• /
• 2010

Taurine, 2-aminoethanesulfonic acid, is an abundant free amino acid present in brain cells and exerts many important biological functions such as anti-convulsant, modulation of neuronal excitability, regulation of learning and memory, anti-aggressiveness and anti-alcoholic effects. In the present study, we investigated to explore whether taurine has any protective actions against oxidative stress-induced damages in neuronal cells. ERK I/II regulates signaling pathways involved in nitric oxide (NO) and reactive oxygen species (ROS) production and plays a role in the regulation of cell growth, and apoptosis. We have found that taurine significantly inhibited AMPA induced cortical depolarization in the Grease Gap assays using rat cortical slices. Taurine also inhibited AMPA-induced neuronal cell damage in MTT assays in the differentiated SH-SY5Y cells. When the neuronal cells were treated with $H_2O_2$, levels of NO were increased; however, taurine pretreatment decreased the NO production induced by $H_2O_2$ to approximately normal levels. Interestingly, taurine treatment stimulated ERK I/II activity in the presence of AMPA or $H_2O_2$, suggesting the potential role of ERK I/II in the neuroprotection of taurine. Taken together, taurine has significant neuroprotective actions against AMPA or $H_2O_2$ induced damages in neuronal cells, possibly via activation of ERK I/II.

• BMB Reports
• /
• v.44 no.11
• /
• pp.730-734
• /
• 2011

Amyloid ${\beta}$-peptide ($A{\beta}$-peptide)-induced oxidative stress is thought to be a critical component of the pathophysiology of Alzheimer's disease (AD). New chalcone derivatives, the Chana series, were recently synthesized from the retrochalcones of licorice. In this study, we investigated the protective effects of the Chana series against neurodegenerative changes in vitro and in vivo. Among the Chana series, Chana 30 showed the highest free radical scavenging activity (90.7%) in the 1,1-diphenyl-2- picrylhydrazyl assay. Chana 30 also protected against $A{\beta}$-induced neural cell injury in vitro. Furthermore, Chana 30 reduced the learning and memory deficits of $A{\beta}_{1-42}$-peptide injected mice. Taken together, these results suggest that Chana 30 may be a promising candidate as a potent therapeutic agent against neurodegenerative diseases.

• Journal of Information Processing Systems
• /
• v.7 no.4
• /
• pp.717-732
• /
• 2011

Classification in biomedical applications is an important task that predicts or classifies an outcome based on a given set of input variables such as diagnostic tests or the symptoms of a patient. Traditionally the classification algorithms would have to digest a stationary set of historical data in order to train up a decision-tree model and the learned model could then be used for testing new samples. However, a new breed of classification called stream-based classification can handle continuous data streams, which are ever evolving, unbound, and unstructured, for instance--biosignal live feeds. These emerging algorithms can potentially be used for real-time classification over biosignal data streams like EEG and ECG, etc. This paper presents a pioneer effort that studies the feasibility of classification algorithms for analyzing biosignals in the forms of infinite data streams. First, a performance comparison is made between traditional and stream-based classification. The results show that accuracy declines intermittently for traditional classification due to the requirement of model re-learning as new data arrives. Second, we show by a simulation that biosignal data streams can be processed with a satisfactory level of performance in terms of accuracy, memory requirement, and speed, by using a collection of stream-mining algorithms called Optimized Very Fast Decision Trees. The algorithms can effectively serve as a corner-stone technology for real-time classification in future biomedical applications.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.5
• /
• pp.1590-1609
• /
• 2021

Laughter is one of the most important nonverbal sound that human generates. It is a means for expressing his emotions. The acoustic and contextual features of this specific sound are different from those of speech and many difficulties arise during their modeling process. During this work, we propose an audio laughter generation system based on unsupervised generative models: the autoencoder (AE) and its variants. This procedure is the association of three main sub-process, (1) the analysis which consist of extracting the log magnitude spectrogram from the laughter database, (2) the generative models training, (3) the synthesis stage which incorporate the involvement of an intermediate mechanism: the vocoder. To improve the synthesis quality, we suggest two hybrid models (LSTM-VAE, GRU-VAE and CNN-VAE) that combine the representation learning capacity of variational autoencoder (VAE) with the temporal modelling ability of a long short-term memory RNN (LSTM) and the CNN ability to learn invariant features. To figure out the performance of our proposed audio laughter generation process, objective evaluation (RMSE) and a perceptual audio quality test (listening test) were conducted. According to these evaluation metrics, we can show that the GRU-VAE outperforms the other VAE models.

• Journal of Korea Multimedia Society
• /
• v.24 no.11
• /
• pp.1552-1559
• /
• 2021

Convolution neural networks (CNNs) show notable performance in image processing and are used as representative core models. CNNs extract and learn features from large amounts of train dataset. In general, it has a structure in which a convolution layer and a fully connected layer are stacked. The core of CNN is the convolution layer. The size of the kernel used for feature extraction and the number that affect the depth of the feature map determine the amount of weight parameters of the CNN that can be learned. These parameters are the main causes of increasing the computational complexity and memory usage of the entire neural network. The most computationally expensive components in CNNs are fully connected and spatial convolution computations. In this paper, we propose a Fourier Convolution Neural Network that performs the operation of the convolution layer in the Fourier domain. We work on modifying and improving the amount of computation by applying the fast fourier transform method. Using the MNIST dataset, the performance was similar to that of the general CNN in terms of accuracy. In terms of operation speed, 7.2% faster operation speed was achieved. An average of 19% faster speed was achieved in experiments using 1024x1024 images and various sizes of kernels.