• Journal of KIBIM
• /
• v.9 no.1
• /
• pp.42-51
• /
• 2019

Maintenance of aging structures has attracted societal attention. Maintenance of the aging structure can be efficiently performed with a digital twin. In order to maintain the structure based on the digital twin, it is required to accurately detect the damage of the structure. Meanwhile, deep learning-based damage detection approaches have shown good performance for detecting damage of structures. However, in order to develop such deep learning-based damage detection approaches, it is necessary to use a large number of data before and after damage, but there is a problem that the amount of data before and after the damage is unbalanced in reality. In order to solve this problem, this study proposed a method based on Generative adversarial network, one of Generative Model, for generating acceleration data usually used for damage detection approaches. As results, it is confirmed that the acceleration data generated by the GAN has a very similar pattern to the acceleration generated by the simulation with structural analysis software. These results show that not only the pattern of the macroscopic data but also the frequency domain of the acceleration data can be reproduced. Therefore, these findings show that the GAN model can analyze complex acceleration data on its own, and it is thought that this data can help training of the deep learning-based damage detection approaches.

• Korean Journal of Applied Biomechanics
• /
• v.28 no.2
• /
• pp.79-85
• /
• 2018

Objective: This study aimed to create a strategic training method to enhance optimal athletic ability using information from 1H to 10H rhythmic units. Method: Top three world class athletes and three national winners of 110-m hurdle finals from the 2010 Daegu International Athletics Competition and 2017 National Athletics Championship, respectively, were selected. To analyze the kinematic variables, Dartfish 9.0 was used for two-dimensional analysis. Results: Regarding the interval time from the start to the finish line, the national athletes took less time during the pure acceleration phase (start to 1H) than the foreign athletes. The horizontal velocity increase was slower after 1H; the national athletes showed a lack of ability to accelerate at the interval phases. Moreover, the hurdle clearance time between phases was longer in the national athletes than in the foreign athletes and lacked consistency. Conclusion: The national athletes lacked the ability to accelerate at the transition, maximum rhythm, rhythm maintenance, and re-acceleration phases and showed a longer hurdle clearance time. If technical improvements and strategic training methods using rhythmic units are applied for hurdling motions, the national athlete's hurdling abilities, performance, and consistency could improve.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.516-518
• /
• 2022

In this paper, we introduce a dependence of number of nodes of hidden-layer in fall detection system using Long Short-Term Memory that can detect falls. Its training is carried out using the parameter theta(θ), which indicates the angle formed by the x, y, and z-axis data for the direction of gravity using a 3-axis acceleration sensor. In its learning, validation is performed and divided into training data and test data in a ratio of 8:2, and training is performed by changing the number of nodes in the hidden layer to increase efficiency. When the number of nodes is 128, the best accuracy is shown with Accuracy = 99.82%, Specificity = 99.58%, and Sensitivity = 100%.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.48 no.2
• /
• pp.147-154
• /
• 2012

In head sea, a ship has mainly the longitudinal motion such as vertical acceleration and pitch. In that case, the motion characteristics of a ship will have changed as the location different from each of place vertically and horizontally on board. The author carried out an experiment about the ship's vertical acceleration and pitch according to the location for the head sea, and analyzed the data with the aid of the statistical and spectral analyzing method to get the motion characteristics of the vessels. The response of vertical acceleration and pitch not deeply depend on the decks vertically, but displayed the relative big different value horizontally even if same deck. The biggest response of vertical acceleration and pitch among the accommodations was shown at scientist room, but the value of it not reached to the minimum requirement of ISO 2631-1 for working on board.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.20 no.4
• /
• pp.412-418
• /
• 2014

In this paper, for better boarding performance and pleasant boarding sensitivity of the ship, comparison and analysis was performed of motion sickness questionnaire with MSI(Motion Sickness Incidence) calculation based on ship motion theory(Strip Method) due to sea condition, incident angle in main sail way, economic speed, and calculation position of the training ship Kaya of Pukyong National University. On theses works, the rougher sea conditions became, the higher total motion sickness rate was occurred. The weights of vertical acceleration and the rates of MSI were higher at the bridge and the accommodation, which were located farther from the center of gravity of the ship. And effects of the vertical acceleration of the ship were increased in rolling then in head sea. In comparison between motion sickness questionnaire with MSI calculation, when the vertical acceleration increased, the motion sickness rate increased. The location to increase vertical acceleration and the location to cause motion sickness were agreed.

• Advances in biomechanics and applications
• /
• v.1 no.1
• /
• pp.57-66
• /
• 2014

The purpose of the study was to investigate the kinematics between the double collar-tie and double underhook Thai Boxing clinching positions. Ten amateur mixed martial arts athletes executed six knee strikes for both clinching positions with their dominant limb directed towards a target. A standard two-dimensional video motion analysis was conducted, and the results showed a statistical significant difference at the hip joint angle and the angular acceleration of the knee and ankle. Within both clinching positions, there was a statistically significant correlation between the hip and knee joint angles, hip and knee angular velocities, and hip angular acceleration. Between both clinching positions, there was a statistically significant correlation at the knee joint angle, knee angular velocity, and hip angular acceleration. This study demonstrates the importance of the hip and knee joint movements in both clinching positions, which implies the applications of strength training and flexibility at these joints for sports performance and injury prevention. It is suggested that future studies analyzing the non-dominant leg are warranted to fully understand the Thai Boxing clinch.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.14 no.6
• /
• pp.190-194
• /
• 2006

This research investigates vehicle structure acceleration and vehicle deformation with RCAR crash test. To investigate vehicle damage characteristics in an individual case, it is possible to RCAR low speed crash test. In this study, two tests were conducted to evaluate difference between RCAR new condition and RCAR old condition. A two large vehicles were subjected to a frontal crash test at a speed of 15km/h with an offset of 40% $10^{\circ}$ angle barrier and flat barrier. The results of the 15km/h with an offset of 40% $10^{\circ}$ angle barrier revealed high acceleration value on the vehicle structure and high repair cost compared to the RCAR 15km/h with an offset of 40% flat barrier. So in order to improve damage characteristics in low speed crash of vehicle structure and body component of the monocoque type passenger vehicles, the end of front side member and front back beam should be designed with optimum level and to supply the end of front side member as a partial condition approx 300mm.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.49 no.10
• /
• pp.883-892
• /
• 2021

This paper provides a process for developing deep learning-based aerial object detection models that can run in realtime on onboard. To improve object detection performance, we pre-process and augment the training data in the training stage. In addition, we perform transfer learning and apply a weighted cross-entropy method to reduce the variations of detection performance for each class. To improve the inference speed, we have generated inference acceleration engines with quantization. Then, we analyze the real-time performance and detection performance on custom aerial image dataset to verify generalization.

• Smart Structures and Systems
• /
• v.28 no.6
• /
• pp.799-810
• /
• 2021

When monitoring the structural integrity of a bridge using data collected through accelerometers, identifying the profile of the load exerted on the bridge from the vehicles passing over it becomes a crucial task. In this study, the speed and location of vehicles on the deck of a bridge is reconfigured using real-time video to implicitly associate the load applied to the bridge with the response from the bridge sensors to develop an image-based deep learning network model. Instead of directly measuring the load that a moving vehicle exerts on the bridge, the intention in the proposed method is to replace the correlation between the movement of vehicles from CCTV images and the corresponding response by the bridge with a neural network model. Given the framework of an input-output-based system identification, CCTV images secured from the bridge and the acceleration measurements from a cantilevered beam are combined during the process of training the neural network model. Since in reality, structural damage cannot be induced in a bridge, the focus of the study is on identifying local changes in parameters by adding mass to a cantilevered beam in the laboratory. The study successfully identified the change in the material parameters in the beam by using the deep-learning neural network model. Also, the method correctly predicted the acceleration response of the beam. The proposed approach can be extended to the structural health monitoring of actual bridges, and its sensitivity to damage can also be improved through optimization of the network training.

• Geomechanics and Engineering
• /
• v.29 no.6
• /
• pp.599-614
• /
• 2022

In this study, the reliability analysis of internal and external stabilities of Reinforced Soil Walls (RSWs) under static and seismic loads are investigated so that it can help the geotechnical engineers to perform the design more realistically. The effect of various variables such as angle of internal soil friction, soil specific gravity, tensile strength of the reinforcements, base friction, surcharge load and finally horizontal earthquake acceleration are examined assuming the variables uncertainties. Also, the correlation coefficient impact between variables, sensitivity analysis, mean change, coefficient of variation and type of probability distribution function were evaluated. In this research, external stability (sliding, overturning and bearing capacity) and internal stability (tensile rupture and pull out) in both static and seismic conditions were investigated. Results of this study indicated sliding as the predominant failure mode in the external stability and reinforcing rupture in the internal stability. First-Order Reliability Method (FORM) are applied to estimate the reliability index (or failure probability) and results are validated using the Monte Carlo Simulation (MCS) method. The results showed among all variables, the internal friction angle and horizontal earthquake acceleration have dominant impact on the both reinforced soil wall internal and external stabilities limit states. Also, the type of probability distribution function affects the reliability index significantly and coefficient of variation of internal friction angle has the greatest influence in the static and seismic limits states compared to the other variables.