• Journal of Biomedical Engineering Research
• /
• v.26 no.4
• /
• pp.243-249
• /
• 2005

The number of people who have lost limbs due to amputation has increased due to various accidents and diseases. Numerous attempts have been made to provide these people with prosthetic devices. These devices are often controlled using myoelectric signals. Although the success of fitting myoelectric signals (EMG) for single device control is apparent, extension of this control to more than one device has been difficult. The lack of success can be attributed to inadequate multifunctional control strategies. Therefore, the objective of this study was to develop multifunctional myoelectric control strategies that can generate a number of output control signals. We demonstrated the feasibility of a neural network classification control method that could generate 12 functions using three EMG channels. The results of evaluating this control strategy suggested that the neural network pattern classification method could be a potential control method to support reliability and convenience in operation. In order to make this artificial neural network control technique a successful control scheme for each amputee who may have different conditions, more investigation of a careful selection of the number of EMG channels, pre-determined contractile motions, and feature values that are estimated from the EMG signals is needed.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.11 no.2
• /
• pp.980-992
• /
• 2019

This paper describes a model test and numerical simulation of a 750-kW-semi-submersible platform wind turbine under several wind and wave conditions for validation of the numerical simulation model. The semi-submersible platform was designed to support the 750-kW-wind turbine class and operate at a water depth of 50 m. The model tests were performed to estimate the performance characteristics of the wind turbine system in the wide tank of the University of Ulsan. Motions and loads of the wind turbine system under the wind and wave conditions were measured and analyzed. The NREL-FAST code was used to simulate the wind turbine system, and the results were compared with those of the test model. The results demonstrate that the numerical simulation captures noticeably the fully coupled floating wind turbine dynamic responses. Also, the model shows a good stability and small responses during waves, wind, and operation of the 750-kW-floating offshore wind turbine.

• Computational Structural Engineering
• /
• v.11 no.1
• /
• pp.179-190
• /
• 1998

A geometrically nonlinear finite element formulation of spatial cable networks is presented using two cable elements. Firstly, derivation procedures of tangent stiffness and mass matrices for the space truss element and the elastic catenary cable element are summarized. The load incremental method based on Newton-Raphson iteration method and the dynamic relaxation method are presented in order to determine the initial static state of cable nets subjected to self-weights and support motions. Furthermore, static non-linear analysis of cable structures under additional live loads are performed based on the initial configuration. Challenging example problems are presented and discussed in order to demonstrate the feasibility of the present finite element method and investigate static nonlinear behaviors of cable nets.

• Journal of Ocean Engineering and Technology
• /
• v.22 no.2
• /
• pp.20-27
• /
• 2008

A moon-pool is a vertical well in a floating barge, drilling ship, or offshore support vessel. In this study, numerical simulation of two-dimensional moon-pool flaw coupled with a ship's motion in waves is carried out using a particle method, the so-called MPS method. The particle method, which is recognized as one of the gridless methods, was developed to investigate nonlinear free-surface motions interacting with structures. The method is more feasible and effective than convectional grid methods in order to solve a flaw field with complicated boundary shapes.

• Nuclear Engineering and Technology
• /
• v.51 no.3
• /
• pp.894-903
• /
• 2019

An approach for collapse risk assessment is proposed to evaluate the vulnerability of electric cabinet in nuclear power plants. The lognormal approaches, namely maximum likelihood estimation and linear regression, are introduced to establish the fragility curves. These two fragility analyses are applied for the numerical models of cabinets considering various boundary conditions, which are expressed by representing restrained and anchored models at the base. The models have been built and verified using the system identification (SI) technique. The fundamental frequency of the electric cabinet is sensitive because of many attached devices. To bypass this complex problem, the average spectral acceleration $S_{\bar{a}}$ in the range of period that cover the first mode period is chosen as an intensity measure on the fragility function. The nonlinear time history analyses for cabinet are conducted using a suite of 40 ground motions. The obtained curves with different approaches are compared, and the variability of risk assessment is evaluated for restrained and anchored models. The fragility curves obtained for anchored model are found to be closer each other, compared to the fragility curves for restrained model. It is also found that the support boundary conditions played a significant role in acceleration response of cabinet.

• Structural Engineering and Mechanics
• /
• v.71 no.4
• /
• pp.445-458
• /
• 2019

This study mainly aims to assess the performance of soil-structure systems designed by direct displacement-based method coupled with strong column-weak beam design concept through various system identification techniques under strong ground motions. To this end, various system identification methods are employed to evaluate the dynamic characteristics of a structure (i.e., modal frequency, system damping, mode shapes, and plastic hinge formation pattern) under a strong seismic excitation considering soil-structure interaction for different site conditions as specified by ASCE 7-10. The scope of the study narrowed down to the code-complying low- to high-rise steel moment resisting frames with various heights (4, 8, 12, 16-story). The comparison of the result of soil-structure systems with fix-based support condition indicates that the modal frequencies of these systems are highly influenced by the structure heights, specifically for the softer soils. This trend is more significant for higher modes of the system which can considerably dominate the response of structures in which the higher modes have more contribution in dynamic response. Amongst all studied modes of the vibration, the damping ratio estimated for the first mode is relatively the closet to the initial assumed damping ratios. Moreover, it was found that fewer plastic hinges are developed in the structure of soil-structure systems with a softer soil which contradicts the general expectation of higher damageability of such structural systems.

• Journal of Biomedical Engineering Research
• /
• v.42 no.4
• /
• pp.186-192
• /
• 2021

In this paper, we evaluated prediction accuracy of Euler angle spectrograph classification method using a convolutional neural networks (CNN) for hand gesture recognition in augmented reality (AR) cognitive rehabilitation system based on Leap Motion Controller (LMC). Hand gesture recognition methods using a conventional support vector machine (SVM) show 91.3% accuracy in multiple motions. In this paper, five hand gestures ("Promise", "Bunny", "Close", "Victory", and "Thumb") are selected and measured 100 times for testing the utility of spectral classification techniques. Validation results for the five hand gestures were able to be correctly predicted 100% of the time, indicating superior recognition accuracy than those of conventional SVM methods. The hand motion recognition using CNN meant to be applied more useful to AR cognitive rehabilitation training systems based on LMC than sign language recognition using SVM.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.1
• /
• pp.56.2-56.2
• /
• 2021

Constraining the structure and thus the fate of giant molecular clouds (GMCs), the primary sites of star formation in galaxies, is crucial to understand the evolution of galaxies themselves. Exploiting the unprecedented sensitivity and angular resolution of the Atacama Large Millimeter/sub-millimeter Array (ALMA), we have measured the spatially-resolved (~ 20 pc resolution) properties of the GMCs in two nearby late-type galaxies, NGC 5806 (SAB(s)b) and NGC 6753 ((R)SA(r)b), as part of the WISDOM project. Although these results are preliminary, we identified ~ 200 resolved GMCs in NGC 5806 within a radius of 500 pc, most within a nuclear ring structure, and ~ 400 resolved GMCs in NGC 6753 within a radius of 2 kpc, most within a flocculent spiral structure. The GMCs of NGC 5806 have similar sizes but slightly higher linewidths than clouds in the Milky Way disc. Because the GMCs also have higher surface densities, the calculated cloud Virial parameters are nevertheless about unity, suggesting that the GMCs of NGC 5806 are in gravitational equilibrium and thus long lived. This is contrary to other WISDOM results on earlier-type galaxies, where large cloud linewidths are likely due to shear associated with the local (circular) orbital motions (rather than the clouds' self-gravity), and the clouds are either marginally or not gravitationally bound. These results support the notion that spheroids alter the dynamical states of clouds (morphological quenching), that are otherwise (i.e. in galaxy discs) fairly homogenous and similar to those of the Milky Way.

• Geomechanics and Engineering
• /
• v.37 no.6
• /
• pp.539-554
• /
• 2024

This study explores development of prediction model for seismic site classification through the integration of machine learning techniques with horizontal-to-vertical spectral ratio (HVSR) methodologies. To improve model accuracy, the research employs outlier detection methods and, synthetic minority over-sampling technique (SMOTE) for data balance, and evaluates using seven machine learning models using seismic data from KiK-net. Notably, light gradient boosting method (LGBM), gradient boosting, and decision tree models exhibit improved performance when coupled with SMOTE, while Multiple linear regression (MLR) and Support vector machine (SVM) models show reduced efficacy. Outlier detection techniques significantly enhance accuracy, particularly for LGBM, gradient boosting, and voting boosting. The ensemble of LGBM with the isolation forest and SMOTE achieves the highest accuracy of 0.91, with LGBM and local outlier factor yielding the highest F1-score of 0.79. Consistently outperforming other models, LGBM proves most efficient for seismic site classification when supported by appropriate preprocessing procedures. These findings show the significance of outlier detection and data balancing for precise seismic soil classification prediction, offering insights and highlighting the potential of machine learning in optimizing site classification accuracy.

• Journal of KIISE:Software and Applications
• /
• v.32 no.3
• /
• pp.215-226
• /
• 2005

Loop invariant code motion (LICM) optimization includes relatively heavy code analyses, thus being not readily applicable to Java Just-In-Time (JIT) compilation where the JIT compilation time is part of the whole running time. 'Classical' LICM optimization first analyzes the code and constructs both the def-use chains and the use-def chains. which are then used for performing code motions. This paper proposes a light-weight LICM algorithm, which requires only the def-use chains of loop invariant code (without use-def chains) by exploiting the fact that the Java virtual machine is based on a stack machine, hence generating code with simpler patterns. We also propose two techniques that allow more code motions than classical LICM techniques. First, unlike previous JIT techniques that uses LICM only in single-path loops for simplicity, we apply LICM to multi-path loops (natural loops) safely for partially redundant code. Secondly, we move loop-invariant, partially-redundant null pointer check code via predication support in Itanium. The proposed techniques were implemented in a JIT compiler for Itanium processor on ORP (Open Runtime Platform) Java virtual machine of Intel. On SPECjvrn98 benchmarks, the proposed technique increases the JIT compilation overhead by the geometric mean of 1.3％, yet it improves the total running time by the geometric mean of 2.2％.