• Journal of the Korea Organic Resources Recycling Association
• /
• v.29 no.4
• /
• pp.47-54
• /
• 2021

In this study, the performance of a honeycomb-type hydrogen oxidation catalyst to remove hydrogen in a hydrogen economy society to secure leaking hydrogen. The Pd/TiO₂ catalyst was prepared based on a liquid phase reduction method that is not exposed to a heat source, and it was showed through H₂-chemisorption analysis that it existed as very small active particles of 2~4 nm. In addition, it was found that the metal dispersion decreased and the active particle size increased as the reduction reaction temperature increased. It was meant that the active metal particle size and the hydrogen oxidation performance were in a proportional correlation, so that it was consistent with the hydrogen oxidation performance reduction result. The prepared catalyst was coated on a support in the form of a honeycomb so that it could be applied to the hydrogen industrial process. When 20 wt% or more of the AS-40 binder was coated, oxidation performance of 90% or more was observed under low-concentration hydrogen conditions. It was showed through SEM analysis that long-term catalytic activity can be expected by enhancing the adhesion strength of the catalyst and preventing catalyst desorption. It is a basic research that can secure safety in a hydrogen society such as gasification, organic resource, and it can be utilized as a system that can respond to unexpected safety accidents in the future.

• Smart Media Journal
• /
• v.11 no.10
• /
• pp.65-75
• /
• 2022

Deep learning is used as a creative tool that could overcome the limitations of existing analysis models and generate various types of results such as text, image, and music. In this paper, we propose a method necessary to preprocess audio data using the Niko's MIDI Pack sound source file as a data set and to generate music using Bi-LSTM. Based on the generated root note, the hidden layers are composed of multi-layers to create a new note suitable for the musical composition, and an attention mechanism is applied to the output gate of the decoder to apply the weight of the factors that affect the data input from the encoder. Setting variables such as loss function and optimization method are applied as parameters for improving the LSTM model. The proposed model is a multi-channel Bi-LSTM with attention that applies notes pitch generated from separating treble clef and bass clef, length of notes, rests, length of rests, and chords to improve the efficiency and prediction of MIDI deep learning process. The results of the learning generate a sound that matches the development of music scale distinct from noise, and we are aiming to contribute to generating a harmonistic stable music.

• Design & Manufacturing
• /
• v.17 no.1
• /
• pp.20-26
• /
• 2023

Lab-on-a-disc is a circular disc shape of cartridge that can be used for blood-based liquid biopsy to diagnose an early stage of cancer. Currently, liquid biopsies are regarded as a time-consuming process, and require sophisticated skills to precisely separate cell-free DNA (cfDNA) and circulating tumor cells (CTCs) floating in the bloodstream for accurate diagnosis. However, by applying the lab-on-a-disc to liquid biopsy, the entire process can be operated automatically. To do so, the lab-on-a-disc should be designed to prevent blood leakage during the centrifugation, transport, and dilution of blood inside the lab-on-a-disc in the process of liquid biopsy. In this study, the main components of lab-on-a-disc for liquid biopsy are fabricated by injection molding for mass production, and ultrasonic welding is employed to ensure the bonding strength between the components. To guarantee accurate ultrasonic welding, the flatness of the components is optimized numerically by using the response surface methodology with four main injection molding processing parameters, including the mold & resin temperatures, the injection speed, and the packing pressure. The 27 times finite element analyses using Moldflow® reveal that the injection time and the packing pressure are the critical factors affecting the flatness of the components with an optimal set of values for all four processing parameters. To further improve the flatness of the lab-on-a-disc components for stable mass production, a quarter-disc shape of lab-on-a-disc with a radius of 75 mm is used instead of a full circular shape of the disc, and this significantly decreases the standard deviation of flatness to 30% due to the reduced overall length of the injection molded components by one-half. Moreover, it is also beneficial to use a quarter disc shape to manage the deviation of flatness under 3 sigma limits.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.36 no.1
• /
• pp.88-92
• /
• 2023

The ferroelectricity in Hf_0.5Zr_0.5O₂ (HZO) thin films is one of the most interesting topics for next-generation nonvolatile memory applications. It is known that a crystallization process is required at a temperature of 400℃ or higher to form an orthorhombic phase that results in the ferroelectric properties of the HZO film. However, to realize the integration of ferroelectric HZO films in the back-end-of-line, it is necessary to reduce the annealing temperature below 400℃. This study aims to comprehensively analyze the ferroelectric properties according to the annealing temperature (350-500℃) and time (1-5 h) using a furnace as a crystallization method for HZO films. As a result, the ferroelectric behaviors of the HZO films were achieved at a temperature of 400℃ or higher regardless of the annealing time. At the annealing temperature of 350℃, the ferroelectric properties appeared only when the annealing time was sufficiently increased (4 h or more). Based on these results, it was experimentally confirmed that the optimization of the annealing temperature and time is very important for the ferroelectric phase crystallization of HZO films and the improvement of their ferroelectric properties.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.4
• /
• pp.213-220
• /
• 2023

This paper presents a model simulating the folding process and collision dynamics of "ddakji", a traditional Korean game played using paper tiles (which are also referred to as ddakji). The model uses two A4 sheets as the base materials for ddakji. The folding process involves a series of boundary conditions that transform the wing part of the paper structure into a twisted configuration. A rigid plate boundary condition is also adopted for squeezing, establishing the shape and stress state of the game-ready ddakji through dynamic relaxation analysis. The gaming process analysis involves a forced displacement of the striking ddakji to a predetermined collision position. Collision analysis then follows at a given speed, with the objective of overturning the struck ddakji--a winning condition. A genetic algorithm-based optimization analysis identifies the optimal collision conditions that result in the overturning of the struck ddakji. For efficiency, the collision analysis is divided into two stages, with the second stage carried out only if the first stage predicts a possible overturn. The fitness function for the genetic algorithm during the first stage is the direction cosine of the struck ddakji, whereas in the second stage, it is the inverse of the speed, thus targeting the lowest overall collision speed. Consequently, this analysis provides optimal collision conditions for various compression thicknesses.

• International journal of advanced smart convergence
• /
• v.12 no.3
• /
• pp.89-103
• /
• 2023

Obstructive sleep apnea (OSA) is one of the most prevalent sleep disorders that can lead to serious consequences, including hypertension and/or cardiovascular diseases, if not treated promptly. Continuous positive airway pressure (CPAP) is widely recognized as the most effective treatment for OSA, which needs the proper titration of airway pressure to achieve the most effective treatment results. However, the process of CPAP titration can be time-consuming and cumbersome. There is a growing importance in predicting personalized CPAP pressure before CPAP treatment. The primary objective of this study was to optimize the CPAP titration process for obstructive sleep apnea patients through EEG feature engineering with machine learning techniques. We aimed to identify and utilize the most critical EEG features to forecast key OSA predictive indicators, ultimately facilitating more precise and personalized CPAP treatment strategies. Here, we analyzed 126 OSA patients' PSG datasets before and after the CPAP treatment. We extracted 29 EEG features to predict the features that have high importance on the OSA prediction index which are AHI and SpO2 by applying the Shapley Additive exPlanation (SHAP) method. Through extracted EEG features, we confirmed the six EEG features that had high importance in predicting AHI and SpO2 using XGBoost, Support Vector Machine regression, and Random Forest Regression. By utilizing the predictive capabilities of EEG-derived features for AHI and SpO2, we can better understand and evaluate the condition of patients undergoing CPAP treatment. The ability to predict these key indicators accurately provides more immediate insight into the patient's sleep quality and potential disturbances. This not only ensures the efficiency of the diagnostic process but also provides more tailored and effective treatment approach. Consequently, the integration of EEG analysis into the sleep study protocol has the potential to revolutionize sleep diagnostics, offering a time-saving, and ultimately more effective evaluation for patients with sleep-related disorders.

• Journal of the Korea Society for Simulation
• /
• v.32 no.3
• /
• pp.33-41
• /
• 2023

As carbon-free has been getting interest, renewable energy sources have been increasing. However, renewable energy is intermittent and variable so it is difficult to predict the produced electrical energy from a renewable energy source. In this study, digital-twin concept is applied to solve difficulties in predicting electrical energy from a renewable energy source. Considering that rotation of wind turbine has high correlation with produced electrical energy, a model which simulates rotation in the drivetrain of a wind turbine is developed. The base of a drivetrain simulation model is set with well-known state equation in mechanical engineering, which simulates the rotating system. Simulation based machine learning is conducted to get unknown parameters which are not provided by manufacturer. The simulation is repeated and parameters in simulation model are corrected after each simulation by optimization algorithm. The trained simulation model is validated with 27 real wind turbine operation data set. The simulation model shows 4.41% error in average compared to real wind turbine operation data set. Finally, it is assessed that the drivetrain simulation model represents the real wind turbine drivetrain system well. It is expected that wind-energy-prediction accuracy would be improved as wind turbine digital twin including the developed drivetrain simulation model is applied.

• Journal of recreation and landscape
• /
• v.8 no.4
• /
• pp.43-51
• /
• 2014

This research is based on a rural village reconstruction business that is a priority under the national support act for rural village vitalization. Allowing for an analysis of the regional and annual classification of business contents as part of the master district implementation plan, this research presents amenity resource applications for the purpose of understanding the business contents and resource status reports. To analyze the utilization of amenity resources in the rural villages' overall development business, a content analysis of the business characteristics and resources of 299 districts was conducted for a seven-year period (2005-2011). Information that included district names, enterprise types, and specifications of a particular business, were coded in Excel, through exhaustive research of the 299 districts. Using this process, a more detailed categorization of seven years of business data, periodic, and regional business contents were defined. As a result of this research, it is apparent that the overall district's facility resources are optimized for the most, and that the environmental management of resources, including animal and plant resources, as well as water resources, is continuously decreasing, as was shown in the annual amenity resource usage transition. The annual amenity resource usage transition data denotes the highest rates in Jun-Ra-Buk-Do and Kyung-Sang-Buk-Do. In summary, this analysis verified the urgent need for diverse amenity resource utilization, research on practical alternatives, and the resource optimization of environmental controls for sustainable development in rural areas.

• Smart Media Journal
• /
• v.11 no.11
• /
• pp.17-24
• /
• 2022

With the spread of smart speakers based on voice recognition technology and deep learning technology, not only non-disabled people, but also the blind or physically handicapped can easily control home appliances such as lights and TVs through voice by linking home network services. This has greatly improved the quality of life. However, in the case of speech-impaired people, it is impossible to use the useful services of the smart speaker because they have inaccurate pronunciation due to articulation or speech disorders. In this paper, we propose a personalized voice classification technique for the speech-impaired to use for some of the functions provided by the smart speaker. The goal of this paper is to increase the recognition rate and accuracy of sentences spoken by speech-impaired people even with a small amount of data and a short learning time so that the service provided by the smart speaker can be actually used. In this paper, data augmentation and one cycle learning rate optimization technique were applied while fine-tuning ResNet18 model. Through an experiment, after recording 10 times for each 30 smart speaker commands, and learning within 3 minutes, the speech classification recognition rate was about 95.2%.

• Journal of the Korean Geotechnical Society
• /
• v.26 no.7
• /
• pp.147-159
• /
• 2010

The settlement prediction is very important in preloading method for a construction site on the soft ground. At the design stage, however, it is hard to predict the settlement exactly due to limitations of the site survey. Most of the settlement prediction is performed by a regression settlement curve based on the field data during construction. In Korea, hyperbolic method has been most commonly used to align the settlement curve with the field data, because of its simplicity and many application cases. The results from hyperbolic method, however, may differ by data selections or data fitting methods. In this study, the analyses using hyperbolic method were performed about the field data of $\bigcirc\bigcirc$ site in Pusan. Two data fitting methods, using an axis transformation or an alternative method which is a direct regression method, were applied with various data groups. If data was used only after the ground water level being stabilized, fitting results using both methods were in good agreement with the measured data. Regardless of the information about the ground water level, the alternative method gives better results with the field data than the method using an axis transformation.