• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.54.2-54
• /
• 2001

In this paper two control strategies, command shaping control and optimal control, which aim to the reduction of the residual vibrations of the payload in a container crane system are investigated. Both control methods are open loop control. Due to unmodeled dynamics of the plant and disturbances like initial sway and wind, some residual sway always exists at the end of trolley movement. Command inputs are designed to achieve the control objectives including minimal residual vibration and robustness in the presence of unmodeled dynamics. Simulation results of various command inputs are compared in terms of arrival time, residual sway angle, robustness, and maximum sway distance during the traveling. Command shaping method provides a more competent tool than optimal control.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.385-398
• /
• 2021

The complexity of deep learning models affects the real-time performance of gesture recognition, thereby limiting the application of gesture recognition algorithms in actual scenarios. Hence, a residual learning neural network based on a deep convolutional neural network is proposed. First, small convolution kernels are used to extract the local details of gesture images. Subsequently, a shallow residual structure is built to share weights, thereby avoiding gradient disappearance or gradient explosion as the network layer deepens; consequently, the difficulty of model optimisation is simplified. Additional convolutional neural networks are used to accelerate the refinement of deep abstract features based on the spatial importance of the gesture feature distribution. Finally, a fully connected cascade softmax classifier is used to complete the gesture recognition. Compared with the dense connection multiplexing feature information network, the proposed algorithm is optimised in feature multiplexing to avoid performance fluctuations caused by feature redundancy. Experimental results from the ISOGD gesture dataset and Gesture dataset prove that the proposed algorithm affords a fast convergence speed and high accuracy.

• Journal of the Semiconductor & Display Technology
• /
• v.16 no.2
• /
• pp.98-105
• /
• 2017

In the display industry, the residual vibration of the transfer robot can increase the process time and cause the breakage of the glass substrate, which is critical to productivity of display manufacturing. In this paper, the natural frequencies of transfer robot are analyzed by finite element method. On the basis of the analyzed data, we investigated the response characteristics of input shaping control with or without the glass presence on the hand of the transfer robot using MATLAB program, and compared with the current response characteristics of input shaping control applied to the industry. Based on this, we suggest an optimal residual vibration control method for the practical application in display industry.

• Korean Journal of Metals and Materials
• /
• v.49 no.6
• /
• pp.468-473
• /
• 2011

Determining the residual stress during casting processes is important for evaluating the mechanical properties and strength of materials and to optimize manufacturing conditions. In this study, we propose a field data interface procedure between FDM and FEM in a 3-dimensional space for analyzing the casting process and structural analysis. The casting process was analyzed using FDM and the data of the temperature distribution were converted into a format suitable for FEM analysis to calculate the thermal stress and safety factor by tightening force. The results of the coupled analysis between FDM and FEM showed that casting residual stress is an important factor in predicting life time and evaluating durability.

• Journal of Information Processing Systems
• /
• v.18 no.1
• /
• pp.12-25
• /
• 2022

In recent years, deep convolutional neural networks have made significant progress in the research of single image super-resolution. However, it is difficult to be applied in practical computing terminals or embedded devices due to a large number of parameters and computational effort. To balance these problems, we propose CSRNet, a lightweight neural network based on channel split residual learning structure, to reconstruct highresolution images from low-resolution images. Lightweight refers to designing a neural network with fewer parameters and a simplified structure for lower memory consumption and faster inference speed. At the same time, it is ensured that the performance of recovering high-resolution images is not degraded. In CSRNet, we reduce the parameters and computation by channel split residual learning. Simultaneously, we propose a double-upsampling network structure to improve the performance of the lightweight super-resolution network and make it easy to train. Finally, we propose a new evaluation metric for the lightweight approaches named 100_FPS. Experiments show that our proposed CSRNet not only speeds up the inference of the neural network and reduces memory consumption, but also performs well on single image super-resolution.

• Proceedings of the KIEE Conference
• /
• 2009.07a
• /
• pp.662_663
• /
• 2009

Motor bus transfer involves the process of transferring a bus that has several critical motors to an alternate source of power when the main normal power source feeding them is interrupted. Bus transfer is a time-critical application in which the transfer progress depends on various parameters such as the type of motor, load on the motor at the time of transfer, inertia of the motor, and the combined open-circuit time constant of various motors present on the bus at the time of transfer. This paper present the result of modeling and simulation of nuclear power motor bus using ETAP(Electrical Transient Analyzing Program) program for motor and motor bus residual voltage decay characteristics.

• Journal of Biosystems Engineering
• /
• v.14 no.1
• /
• pp.8-15
• /
• 1989

Agricultural materials do not react in a purely elastic manner, and their responses when subjected to stress and strain are appeared from a combination of elastic and viscous behavior. Various researchers have studied the mechanical and rheological properties of the many agricultural materials, but those properties are available mostly foreign varieties of agricultural products. Rheological properties of rice seedlings become important to formulate the principles governing their mechanical behavior. The objectives of this study were to experimentally determine the stress relaxation properties of rice seedlings such as three Japonica-type and one Indica ${\times}$ Japonica hybrid in the transplanting age. The results of this study are summarized as follows; 1. The stress relaxation behavior could be described by the generalized Maxwell model. 2. The phenomenon of stress relaxation happened abruptly just after loading and this phenomenon weakened with the loading time lapsed. 3. With increase of the initial stress, the stress relaxation intensity and residual stress increased, while the relaxation time was constant with increased, while the relaxation time was constant with increase of the level of initial stress. 4. With increase of loading rate, the stress relaxation intensity increased, while the relaxation time and residual stress decreased.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2000.11a
• /
• pp.783-786
• /
• 2000

For the development of a ultra-precision CMC polishing system including on-machine measurement system, we study a corrective polishing algorithm. We calculated unit removal profiles for various polishing tools and polishing tool positions. Using these results we simulate the corrective polishing process based on dwell time control. We calculate dwell time distributions and residual error of the polishing simulation method and the FFT calculation method. We got good dwell time distributions and small residual when we used the FFT calculation method. This results will be used for the optimization of corrective polishing process.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.19 no.11
• /
• pp.1033-1036
• /
• 2006

We have investigated electro-optical characteristics in three kinds of twisted nematic (TN) cells on the polyimide surface. The threshold voltage and the response time of thermal stressed TN cells were same that of no thermal stressed TN cells. There were little change of value in these TN cells. On the other hand, transmittance of no thermal stressed TN cells were better than that of thermal stressed TN cells. Transmittances of TN cells on the polyimide surface decreased by increasing thermal stress time. Moreover, the residual DC of the thermal stressed TN cells increased as increasing thermal stress temperature and time. Therefore, thermal stability of TN cells were decreased gradually by giving high thermal stress for a long time.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2017.05a
• /
• pp.1000-1002
• /
• 2017

Here, the unsteady Navier-Stokes solver has been developed using implicit dual time stepping method. The implicit dual time stepping method introduced the pseudo time step for solving the new residual including the steady state residual and real time derivative. For the validation of code, Stokes 2nd problem, the laminar flow on the oscillating flat plate was selected and compare the calculating results with analytic solutions. The calculating velocity profile and skin friction has a good agreement with analytic solutions.