• 소성∙가공
• /
• 제21권7호
• /
• pp.426-431
• /
• 2012

The work roll of a Sendzimir mill has a small diameter in comparison to its length, so it is easily deformed by the rolling pressure. It also has a complex back up roll system, so it is difficult to analyze the roll deformation. For this reason in Part I we have developed a model which predicts the radial displacement of the roll. In this paper, we apply the model to a Sendzimir mill and propose a new model for the prediction of the deformed roll profile in a Sendzimir mill. The prediction accuracy of the new model is demonstrated through comparison of the predictions from the FE model.

• 소성∙가공
• /
• 제7권3호
• /
• pp.282-290
• /
• 1998

In bulk metal forming processes prediction of tool life is very important for saving production cost and achieving good material properties. Generally the service life of tools in metal forming process is limited to a large extent by wear, fracture and plastic deformation of tools. In case of hot and warm forging processes tool life depends on wear over 70%. In this study finite element analyses are con-ducted to warm and hot forging by adopting suggested wear model. By comparison of simulation and eal profile of die suggested wear model. By comparison of simulation and real profile of die suggested model is verified.

• Journal of Information Processing Systems
• /
• 제7권4호
• /
• pp.635-652
• /
• 2011

In ubiquitous environments, many applications need to process data with time and space dimensions. Because of this, there is growing attention not only on gathering spatiotemporal data in ubiquitous environments, but also on processing such data in databases. In order to obtain the full benefits from spatiotemporal data, we need a data model that naturally expresses the properties of spatiotemporal data. In this paper, we introduce three spatiotemporal data models extended from temporal data models. The main goal of this paper is to determine which data model is less complex in the spatiotemporal context. To this end, we compare their query languages in the complexity aspect because the complexity of a query language is tightly coupled with its underlying data model. Throughout our investigations, we show that it is important to intertwine space and time dimensions and keep one-to-one correspondence between an object in the real world and a tuple in a database in order to naturally express queries in ubiquitous applications.

• 한국원자력학회:학술대회논문집
• /
• 한국원자력학회 1999년도 추계학술발표회요약집
• /
• pp.173.2-173
• /
• 1999

• 천문학회보
• /
• 제37권2호
• /
• pp.124.1-124.1
• /
• 2012

We have made a comparison of the WSA-ENLIL CME propagation model with three cone types and an empirical model using 29 halo CMEs from 2001 to 2002. These halo CMEs have cone model parameters from Michalek et al. (2007) as well as their associated interplanetary (IP) shocks. For this study we consider three different cone models (an asymmetric cone model, an ice-cream cone model and an elliptical cone model) to determine CME cone parameters (radial velocity, angular width and source location), which are used for input parameters of the WSA-ENLIL CME propagation model. The mean absolute error (MAE) of the arrival times at the Earth for the elliptical cone model is 10 hours, which is about 2 hours smaller than those of the other models. However, this value is still larger than that (8.7 hours) of an empirical model by Kim et al. (2007). We are investigating several possibilities on relatively large errors of the WSA-ENLIL cone model, which may be caused by CME-CME interaction, background solar wind speed, and/or CME density enhancement.

• 천문학회보
• /
• 제37권1호
• /
• pp.92.1-92.1
• /
• 2012

In this work we have examined the performance of the WSA/ENLIL cone model provided by Community Coordinated Modeling Center (CCMC). The WSA/ENLIL model simulates the propagation of coronal mass ejections (CMEs) from the Sun into the heliosphere. We estimate the shock arrival times at the Earth using 29 halo CMEs from 2001 to 2002. These halo CMEs have cone model parameters from Michalek et al. (2007) as well as their associated interplanetary (IP) shocks. We make a comparison between CME arrival times by the WSA/ENLIL cone model and IP shock observations. For the WSA/ENLIL cone model, the root mean square(RMS) error is about 13 hours and the mean absolute error(MAE) is approximately 10.4 hours. We compared these estimates with those of the empirical model by Kim et al.(2007). For the empirical model, the RMS and MAE errors are about 10.2 hours and 8.7 hours, respectively. We are investigating several possibilities on relatively large errors of the WSA/ENLIL cone model, which may be caused by cone model velocities, CME density enhancement factor, or CME-CME interaction.

• Smart Structures and Systems
• /
• 제24권3호
• /
• pp.361-377
• /
• 2019

The stiffness of shape memory alloy (SMA) spring while in actuation is represented by an empirical model that is derived from the logistic differential equation. This model correlates the stiffness to the alloy temperature and the functionality of SMA spring as active variable stiffness actuator (VSA) is analyzed based on factors that are the input conditions (activation current, duty cycle and excitation frequency) and operating conditions (pre-stress and mechanical connection). The model parameters are estimated by adopting the nonlinear least square method, henceforth, the model is validated experimentally. The average correlation factor of 0.95 between the model response and experimental results validates the proposed model. In furtherance, the justification is augmented from the comparison with existing stiffness models (logistic curve model and polynomial model). The important distinction from several observations regarding the comparison of the model prediction with the experimental states that it is more superior, flexible and adaptable than the existing. The nature of stiffness variation in the SMA spring is assessed also from the Dynamic Mechanical Thermal Analysis (DMTA), which as well proves the proposal. This model advances the ability to use SMA integrated mechanism for enhanced variable stiffness actuation. The investigation proves that the stiffness of SMA spring may be altered under controlled conditions.

• Biotechnology and Bioprocess Engineering:BBE
• /
• 제5권1호
• /
• pp.65-70
• /
• 2000

An approximated analytical solution of mathematical model for the three phase fluidized bed bioreactor (TFBBR) was proposed using the linearization technique to describe oxygen utilization rate in wastewater treatment. The validation of the model was done in comparison with the experimental results. Satisfactory agreement was obtained in the comparison of approximated analytical solution and numerical solution in the oxygen concentration profile of a TFBBR. The approximated solutions for three modes of the liquid phase flow were compared. The proposed model was able to predict the biomass concentration, dissolved oxygen concentration the height of efficient column, and the removal efficiency.

• Journal of Mechanical Science and Technology
• /
• 제15권9호
• /
• pp.1319-1327
• /
• 2001

The results of stand and field testing of a combustion chamber for a heavy-duty 150 MW gas turbine are discussed. The model represented one of 14 identical segments of a tubular multican combustor constructed 1:1 scale. The model experiments were executed at a lower pressure than that in a real gas turbine. Combustion efficiency, pressure loss factor, pattern factor, liner wall temperature, flame radiation, fluctuating pressure and NOx emission were measured at partial and full loads for both model and on-site testing. The comparison of these items in the stand and field test results led to has the development of a method of calculation and the improvement of gas turbine combustors.

• 반도체디스플레이기술학회지
• /
• 제21권3호
• /
• pp.63-68
• /
• 2022

In this study, we produced defect detection models using parallel convolution neural networks. If convolution neural networks are constructed parallel type, the model's detection accuracy will increase and detection time will decrease. We produced parallel-type defect detection models using 4 types of convolutional algorithms. The performance of models was evaluated using evaluation indicators. The model's performance is detection accuracy and detection time. We compared the performance of each parallel model. The detection accuracy of the model using AlexNet is 97 % and the detection time is 0.3 seconds. We confirmed that when AlexNet algorithm is constructed parallel type, the model has the highest performance.