• Transactions of the Korean Society of Automotive Engineers
• /
• v.4 no.4
• /
• pp.107-117
• /
• 1996

The shift feeling, a driver experiences during gear shifting, is a major factor in manual transmission quality. Recently, the shift feeling has becoming more severe every year in proportion to the higher torque and revolution speed of today's automotive engine. In this paper, first, the diagram of a relation between cone angle, sleeve chamfer angle, and friction coefficient of ring is investigated for easy design of Synchromesh system. And then, methodology to solve the shift problems such as clashing noise and rough shift, ect. by analyzing the synchronization procedure in sequence and by investigating the shift waveform if presented.

• Journal of KIISE:Software and Applications
• /
• v.36 no.11
• /
• pp.936-946
• /
• 2009

A graph cuts method has recently attracted a lot of attentions for image segmentation, as it can globally minimize energy functions composed of data term that reflects how each pixel fits into prior information for each class and smoothness term that penalizes discontinuities between neighboring pixels. In previous approaches to graph cuts-based automatic image segmentation, GMM(Gaussian mixture models) is generally used, and means and covariance matrixes calculated by EM algorithm were used as prior information for each cluster. However, it is practicable only for clusters with a hyper-spherical or hyper-ellipsoidal shape, as the cluster was represented based on the covariance matrix centered on the mean. For arbitrary-shaped clusters, this paper proposes graph cuts-based image segmentation using mean shift analysis. As a prior information to estimate the data term, we use the set of mean trajectories toward each mode from initial means randomly selected in $L^*u^*{\upsilon}^*$ color space. Since the mean shift procedure requires many computational times, we transform features in continuous feature space into 3D discrete grid, and use 3D kernel based on the first moment in the grid, which are needed to move the means to modes. In the experiments, we investigate the problems of mean shift-based and normalized cuts-based image segmentation methods that are recently popular methods, and the proposed method showed better performance than previous two methods and graph cuts-based automatic image segmentation using GMM on Berkeley segmentation dataset.

• Journal of Drive and Control
• /
• v.17 no.3
• /
• pp.47-56
• /
• 2020

As a shift control of automatic transmission was managed with the electronic control unit (ECU), shift quality which is a measure of shift shock during gear change has markedly improved. However, the initial clutch pressure control of the clutch filling phase should continue to rely on the predetermined control input since the input and output speeds are unchanged until the shifting process attains the inertia phase. It is critical to minimize the clutch response time and control the clutch pressure accurately at the end of clutch fill to achieve quick shift response and smoothness. Advanced transmission companies have adopted an auto calibration method which establishes the databases for the clutch piston fill-up attributes and the frictional characteristics of the disks. In this study, a distinctive auto calibration algorithm for forklift transmission under development is proposed and verified with the real-vehicle test. The experimental calibration results showed consistent turbine dynamics at the initial stage of shifts with the properly calibrated clutch-fill control parameters. By using this technique, it is necessary to finalize the shift control for the various operation conditions.

• Korean Journal of Computational Design and Engineering
• /
• v.12 no.1
• /
• pp.58-73
• /
• 2007

This paper described a method for the construction of a surface through a set of nonuniform scattered points. When the shift vectors of some points as constraints on the original surface are given, those of the other points should be computed to make the new surface. To keep up the look-see and smoothness with the original surfaces, the proper relationship should be formulated between the shifts of the constraint points and those of the other points. Vector fields for 3 dimensional shift of a point on the surface are made based in the constraint shifts. Vector fields for 3 dimensional shift of a point on the surface are made based on the constraint shifts. Multilevel B-spline approximation technique was used to construct the vector field. The technique uses coarse-to-fine hierarchy of control lattices. The developed method was applied to shoe sole design system especially for grading. Using this system, a shoe sole can be modified effectively.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.31A no.6
• /
• pp.107-116
• /
• 1994

In this paper, The electrical properties of MOS capacitors and transistoras with gate of in-situ doped amorphous Si and poly Si doped by POCI$_3$. Under constant current F-N stress, MOS capacitors with in-situ doped amorphous Si gate have shown the best resistance to degradation in reliabilty properties such as increase of leakage current, shift of gate voltage (V$_{g}$). shift of flat band voltage (V$_{fb}$) and charge to breakdown(Q$_{bd}$). Also, MOSFETs with in-situ doped amorphous Si gate have shown to have less degradation in transistor properties such as threshold voltage, transconductance and drain current. These improvements observed in MOS devices with in-situ doped amorphous Si gate is attributed to less local thinning spots at the gate/SiO$_2$ interface, caused by the large grain size and the smoothness of the surface at the gate/SiO$_2$ interface.

• Geophysics and Geophysical Exploration
• /
• v.6 no.4
• /
• pp.199-206
• /
• 2003

Characteristics of the static shift are discussed by comparing the three-dimensional MT inversion with/without static shift parameterization. The galvanic distortion by small-scale shallow feature often leads severe distortion in inverted resistivity structures. The new inversion algorithm is applied to four numerical data sets contaminated by different amount of static shift. In real field data interpretations, we generally do not have any a-priori information about how much the data contains the static shift. In this study, we developed an algorithm for finding both Lagrangian multiplier for smoothness and the trade-off parameter for static shift, simultaneously in 3-D MT inversion. Applications of this inversion routine for the numerical data sets showed quite reasonable estimation of static shift parameters without any a-priori information. The inversion scheme is successfully applied to all the four data sets, even when the static shift does not obey the Gaussian distribution. Allowing the static shift parameters have non-zero degree of freedom to the inversion, we could get more accurate block resistivities as well as static shifts in the data. When inversion does not consider the static shift as inversion parameters (conventional MT inversion), the block resistivities on the surface are modified considerably to match possible static shift. The inhomogeneous blocks on the surface can generate the static shift at low frequencies. By those mechanisms, the conventional 3-D MT inversion can reconstruct the resistivity structures to some extent in the deeper parts even when moderate static shifts are in the data. As frequency increased, however, the galvanic distortion is not frequency independent any more, and thus the conventional inversion failed to fit the apparent resistivity and phase, especially when strong static shift is added. Even in such case, however, reasonable estimation of block resistivity as well as static shift parameters were obtained by 3-D MT inversion with static shift parameterization.

• Journal of Korea Water Resources Association
• /
• v.53 no.12
• /
• pp.1159-1172
• /
• 2020

This study applied deep convolution neural network based on U-Net and SegNet using long period weather radar data to very short-term rainfall prediction. And the results were compared and evaluated with the translation model. For training and validation of deep neural network, Mt. Gwanak and Mt. Gwangdeoksan radar data were collected from 2010 to 2016 and converted to a gray-scale image file in an HDF5 format with a 1km spatial resolution. The deep neural network model was trained to predict precipitation after 10 minutes by using the four consecutive radar image data, and the recursive method of repeating forecasts was applied to carry out lead time 60 minutes with the pretrained deep neural network model. To evaluate the performance of deep neural network prediction model, 24 rain cases in 2017 were forecast for rainfall up to 60 minutes in advance. As a result of evaluating the predicted performance by calculating the mean absolute error (MAE) and critical success index (CSI) at the threshold of 0.1, 1, and 5 mm/hr, the deep neural network model showed better performance in the case of rainfall threshold of 0.1, 1 mm/hr in terms of MAE, and showed better performance than the translation model for lead time 50 minutes in terms of CSI. In particular, although the deep neural network prediction model performed generally better than the translation model for weak rainfall of 5 mm/hr or less, the deep neural network prediction model had limitations in predicting distinct precipitation characteristics of high intensity as a result of the evaluation of threshold of 5 mm/hr. The longer lead time, the spatial smoothness increase with lead time thereby reducing the accuracy of rainfall prediction The translation model turned out to be superior in predicting the exceedance of higher intensity thresholds (> 5 mm/hr) because it preserves distinct precipitation characteristics, but the rainfall position tends to shift incorrectly. This study are expected to be helpful for the improvement of radar rainfall prediction model using deep neural networks in the future. In addition, the massive weather radar data established in this study will be provided through open repositories for future use in subsequent studies.