• Transactions of Materials Processing
• /
• v.30 no.4
• /
• pp.186-194
• /
• 2021

Finite element simulation is a widely applied method for practical purpose in various metal forming process. However, in the simulation of elasto-plastic behavior of porous material or in crystal plasticity coupled multi-scale simulation, it requires much calculation time, which is a limitation in its application in practical situations. A machine learning model that directly outputs the constitutive equation without iterative calculations would greatly reduce the calculation time of the simulation. In this study, we examined the possibility of artificial intelligence based constitutive equation with the input of existing state variables and current velocity filed. To introduce the methodology, we described the process of obtaining the training data, machine learning process and the coupling of machine learning model with commercial software DEFROM^TM, as a preliminary study, via rigid plastic finite element simulation.

• The Journal of Korean Physical Therapy
• /
• v.21 no.3
• /
• pp.33-40
• /
• 2009

Purpose: This study examined the effects of a virtual reality-based exercise program on the functional recovery of balance and gait in chronic stroke subjects. Methods: A total of 42 chronic stroke patients were enrolled in this study. The participants were allocated randomly to 2 groups: a VR (n=22) and control group (n=20). Both groups received treadmill training for 3 sessions (10 minutes each), 30 minutes per week over a 6 week period. The VR group practiced additional virtual reality programs consisting of 3 programs for 10 minutes each. The data was analyzed using a paired t-test and independent t-test to determine the statistical significance. Results: The virtual reality-based exercise group showed significant increases in gait velocity, cadence and stride length compared to the control group (p<0.05). However, there were no significant differences in static balance. Conclusion: These results support the perceived benefits of exercise programs that incorporate virtual reality to augment the balance and ambulation of stroke patients. Therefore, virtual reality is feasible and suitable for stroke patients

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.6
• /
• pp.1120-1127
• /
• 2008

This paper presents implementation of localization system using UWB (Ultra-Wide Band) sensors and its experimental results along with development of context-aware services. In order for precise measurement of position, we experimented various conditions of pitch angles, yaw angles, number of sensors, height of tags along with measuring errors at each installation. As an application examples of the location tracking system, we developed an intelligent health training management system based on context-aware technology. The system provides appropriate training schedule to a trainee by recognizing position of the trainee and current status of gymnastic equipments and note the usage of the equipment through a personal digital assistant (PDA). Error compensation on position data and moving direction of the trainee was necessary for context-aware service. Hence, we proposed an error compensation algorithm using velocity of the trainee. Experimental results showed that proposed algorithm had made error data reduce by 30% comparing with the data without applying the algorithm.

• Korean Journal of Applied Biomechanics
• /
• v.21 no.3
• /
• pp.383-390
• /
• 2011

In this study, 26 normal subjects were studied to compare the biomechanical Analysis of Lower Limbs on Speed of Nordic Walking. The biomechanical variables were determined by performing three-dimensional gait analysis, and the measurements items were spatial and temporal parameters; vertical ground reaction force; and moments of the hip, knee, and ankle joints. The purpose of this study based on the speed of Nordic Walking to the vertical ground reaction force and joint moments of each were analyzed. Nordic Walking with poles while being whether this weight is reduced to load, not the improvement of muscle activity by identify Nordic walking is to allow efficient. The results of the analysis were follows. The spatial parameters of step length, stride length significantly increased with increase in velocity(p<0.001). The temporal parameters of step time, stride time, the duration of double support use, and the duration of single support use also significantly decreased with increase in velocity(p<0.001), but cadence significantly increased(p<0.01). Analysis of the changes in ground reaction force revealed that vertical ground reaction force significantly increased at the initial contact and the terminal stance and decreased at the mid stance with increase in velocity(p<0.001). Moments of the hip and knee joints significantly in creased with increase in velocity whereas that of the ankle joint did not. Gait analysis revealed that weight-bearing decreased and moments of the hip and knee joints increased with increase in velocity(p<0.01). The results of this study may help people perform Nordic walking efficiently and Nordic walking can be used in the gait training of people with an abnormal gait.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.3
• /
• pp.242-247
• /
• 2015

In this study, we propose the design of Radial Basis Function Neural Network(RBFNN) classifier in order to classify between precipitation and non-precipitation echo. The characteristics of meteorological radar data is analyzed for classifying precipitation and non-precipitation echo. Input variables is selected as DZ, SDZ, VGZ, SPN, DZ_FR, VR by performing pre-processing of UF data based on the characteristics analysis and these are composed of training and test data. Finally, QC data being used in Korea Meteorological Administration is applied to compare with the performance results of proposed classifier.

• Geophysics and Geophysical Exploration
• /
• v.25 no.4
• /
• pp.227-241
• /
• 2022

Full waveform inversion (FWI) in the field of seismic data processing is an inversion technique that is used to estimate the velocity model of the subsurface for oil and gas exploration. Recently, deep learning (DL) technology has been increasingly used for seismic data processing, and its combination with FWI has attracted remarkable research efforts. For example, DL-based data processing techniques have been utilized for preprocessing input data for FWI, enabling the direct implementation of FWI through DL technology. DL-based FWI can be divided into the following methods: pure data-based, physics-based neural network, encoder-decoder, reparameterized FWI, and physics-informed neural network. In this review, we describe the theory and characteristics of the methods by systematizing them in the order of advancements. In the early days of DL-based FWI, the DL model predicted the velocity model by preparing a large training data set to adopt faithfully the basic principles of data science and apply a pure data-based prediction model. The current research trend is to supplement the shortcomings of the pure data-based approach using the loss function consisting of seismic data or physical information from the wave equation itself in deep neural networks. Based on these developments, DL-based FWI has evolved to not require a large amount of learning data, alleviating the cycle-skipping problem, which is an intrinsic limitation of FWI, and reducing computation times dramatically. The value of DL-based FWI is expected to increase continually in the processing of seismic data.

• The Journal of Korean society of community based occupational therapy
• /
• v.10 no.1
• /
• pp.51-61
• /
• 2020

Objectives : The purpose of this study is to investigate the effect of object size and reaching distance on kinematic factors of the upper limb while performing arm reaching for normal subjects. Methods : The subjects of this study were 30 university students who were in D university in Busan, and the measuring tool was CMS-70P(Zebris Medizintechnik Gmbh, Germany), a three-dimensional motion analyzer. The task had six conditions. The average velocity of motion, average acceleration, maximum velocity, and the velocity definite number of movements were measured according to changes in object size(2cm, 10cm) and reaching distance(15%, 37.5%, 60%) when they performed arm reaching. The general characteristics of the subject were technical statistics. One-way ANOVA measurement was used to compare variables when the arm reaching task was performed from two object sizes to three reaching distance, and the post-test was conducted with Tukey test. In addition, an independent t-test was used to analyze the kinematic differences according to the two object sizes at three reaching distances. A two-way ANOVA measurement (3×2 Two-way ANOVA measurement) was performed to identify the interaction of the reaching distance(15%, 37.5%, 60%) and the object size(2cm, 10cm). The statistical significance level α was set to .05. Results : When the size of the object increased, the velocity and maximum velocity also increased, but the definite number of velocity decreased. When the reaching distance increased, the velocity and maximum velocity increased, whereas the definite number of velocity decreased. Conclusion : The clinical significance of this study could be utilized as the baseline data for grading object size and reaching distances when the reaching training is implemented for patients whose central nervous system was damaged.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.56 no.2
• /
• pp.172-182
• /
• 2020

In order to offer specific information needed to assist in operation of a ship with same type rudder through evaluating the maneuverability of training ship A-Ra with flapped rudder, sea trials based full scale for turning test, zig-zag test with rudder angle 10° and 20°, and spiral test at service condition were carried out on starboard and port sides around Jeju Island according to the standards of maneuverability of IMO. As a result, the angular velocity of port turn was higher than that of starboard turn. Therefore, the size of turning circle was longer on the starboard side. In addition, variation of the transfer due to various factors was more stable than those of the others. In the Z-test results, the mean of 1st and 2nd overshoot angles were 9.8°, 6.3° and 15.3°, 9.2° respectively when the port and starboard was 10°; the 1st overshoot angle were 18°, 13.7° when using 20°. Her maneuverability index T' and K' can be easily determined by using a computer with the data obtained from Z-test where K' and T' are dimensionless constants representing turning ability and responsiveness to the helm, respectively. In the Z-test under flap rudder angle 10°, the obtained K' value covered the range of 2.37-2.87 and T' was 1.74-3.45. Under the flap rudder angle 20°, K' and T' value showed 1.43-1.63, 1.0-1.73, respectively. In the spiral test, the loop width was unstable at +0.3° and -0.5°-0.9° around the midship of flap rudder. As a result, course stability was comparatively good. From the sea trial results, training ship ARA met the present criterion in the standards of maneuverability of IMO.

• Journal of the Korean Society for Precision Engineering
• /
• v.33 no.9
• /
• pp.753-760
• /
• 2016

This study was undertaken to develop a bed-type cycling system of lower limbs for rehabilitation. This system consists of two modes of cycling: active and passive. Different velocity and loads are provided for improving the muscle function recovery and increasing the muscular strength. To analyze the muscle activity pattern, we measured muscular activity of lower limbs in the rectus femoris (RF), biceps femoris (BF), tibialis anterior (TA), medial gastrocnemius (MG), and soleus (SOL), while cycling in the supine position, and based on the pedaling direction. A total of 18 young and 23 elderly, healthy subjects participated in this study. Muscle activity of MG muscles was significantly different in the two age groups. This study could provide the reference data to develop cycling exercises for lower limbs during rehabilitation of the elderly patients.

• Journal of The Korean Society of Integrative Medicine
• /
• v.7 no.1
• /
• pp.1-8
• /
• 2019

Purpose : The purpose of this study was to examine the correlation between static and dynamic balance according to the virtual reality-based squat and conventional squat exercise. Methods : Twenty four participants were randomly assigned to the virtual reality-based squat (VRS) group (n=12) or conventional squat (CS) group (n=12). The static balance (C90 area, C90 angle, trace length, sway average velocity) and dynamic balance (forward, rearward, leftward, rightward) were measured using a force plate by BT4. The VRS group used the virtual reality system during 4 weeks, while the CS group underwent classical squat training. Independent t-test was used to test the homogeneity of the general characteristics of the subjects. The collected data was analyzed using the paired t-test for static and dynamic balance comparisons before and after exercise in both groups and Pearson's test for the correlation between static and dynamic balance according to the measured time. The significance level was set to 0.05. Results : There was no significant correlation between group and static and dynamic balance related variables (p>.05). There was a significant correlation between measurement time and static and dynamic balance related variables (p<.05). According to the measurement time, the static balance parameter C90 area in the VRS group after exercise was significantly decreased (p<.05). The values of forward, leftward and rightward in the VRS group were significantly increased after exercise (p<.05). Conclusion : It is suggested that 20 normal healthy adult men and women who have normal balance ability can improve their ability to control their posture by improving the balance ability when applying virtual reality-based squat exercise.