• The Transactions of the Korean Institute of Power Electronics
• /
• v.26 no.5
• /
• pp.364-371
• /
• 2021

A magnetic flux saturation model of Synchronous Reluctance Motors (SynRMs) and a parameter estimation method are proposed at standstill. The proposed magnetic flux model includes the nonlinear relationship between the current and the magnetic flux for self-saturation and cross-saturation. Voltage is injected at standstill to estimate the magnetic flux saturation model. Voltages are injected into the d-axis and q-axis to obtain data on self-saturation. Subsequently, voltages are simultaneously injected into the d-q axis to obtain data on cross-saturation. On the basis of the measured current and the calculated magnetic flux, the parameters of the proposed model are estimated using the least square method (LSM). Simulation and experiment were performed on a 1.5-kW SynRM to verify the proposed method. The proposed model can be used to create a high-efficiency operation table, a sensorless algorithm, and a current controller to improve the control performance of a motor.

• Journal of information and communication convergence engineering
• /
• v.19 no.2
• /
• pp.79-83
• /
• 2021

This paper proposes a novel image classification method based on few-shot learning, which is mainly used to solve model overfitting and non-convergence in image classification tasks of small datasets and improve the accuracy of classification. This method uses model structure optimization to extend the basic convolutional neural network (CNN) model and extracts more image features by adding convolutional layers, thereby improving the classification accuracy. We incorporated certain measures to improve the performance of the model. First, we used general methods such as setting a lower learning rate and shuffling to promote the rapid convergence of the model. Second, we used the data expansion technology to preprocess small datasets to increase the number of training data sets and suppress over-fitting. We applied the model to 10 monkey species and achieved outstanding performances. Experiments indicated that our proposed method achieved an accuracy of 87.92%, which is 26.1% higher than that of the traditional CNN method and 1.1% higher than that of the deep convolutional neural network ResNet50.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.1
• /
• pp.113-117
• /
• 2023

As sputter equipment becomes more complex, it becomes increasingly difficult to understand the parameters that affect the thickness uniformity of thin metal film deposited by sputter. To address this issue, we verified a deep learning model that can predict complex relationships. Specifically, we trained the model to predict the height of 36 magnets based on the thickness of the material, using Support Vector Machine (SVM), Multilayer Perceptron (MLP), 1D-Convolutional Neural Network (1D-CNN), and 2D-Convolutional Neural Network (2D-CNN) algorithms. After evaluating each model, we found that the MLP model exhibited the best performance, especially when the dataset was constructed regardless of the thin film material. In conclusion, our study suggests that it is possible to predict the sputter equipment source using film thickness data through a deep learning model, which makes it easier to understand the relationship between film thickness and sputter equipment.

• Journal of the Korea Convergence Society
• /
• v.5 no.4
• /
• pp.163-169
• /
• 2014

A stable neural network control scheme for unknown non-linear systems is developed in this paper. While the control variable is optimised to minimize the performance index, convergence of the index is guaranteed asymptotically stable by a Lyapnov control law. The optimization is achieved using a gradient descent searching algorithm and is consequently slow. A fast convergence algorithm using an adaptive learning rate is employed to speed up the convergence. Application of the stable control to a single input single output (SISO) non-linear system is simulated. The satisfactory control performance is obtained.

• Proceedings of the IEEK Conference
• /
• 2003.11c
• /
• pp.93-96
• /
• 2003

In this paper, we analyze the concept of network convergence and characteristics of Broadband convergence Network(BcN) from the service perspective where convergence would appear. We suggest the evolutionary 3 stage steps of BcN as an ideal network strategy when considering present network situation. We also propose the features of several representative services that are expected to emerge out of network convergence.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.04a
• /
• pp.1046-1047
• /
• 2012

산업화 시대에서 지식정보화 사회로, 현재는 융합 사회로 진화되고 있다. 하지만 교육은 현재의 사회를 반영하지 못하고 아직도 단일 교과 중심의 교육이 이루어 지고 있다. 이에 고등학교 교육에 융합교육 모형을 개발하고자 한다.

• Journal of the Korea Convergence Society
• /
• v.8 no.10
• /
• pp.215-223
• /
• 2017

This article has tried to analyse the effect of the corporate earning return tax empirically through analysis on the impact of previous internal reservation on the dividends rate of the current year. In addition to this, this article has tried to the effectiveness of government policies with leverage ratio as a moderating variable. Moreover, DRF and GBM model were used to see the effect again. As a result of the actual proof analysis, OCF, ROE, FOR have a significance level of 99% in model1, model2, model3. However, ADV and MSE has appeared not to be meaningful in all models. In the result of DRF and GBM model for convergence was higher than GBM in depth and leaves. However, when it comes to a model explaining capability, GBM high than DRF. The further study will be required to examine the effect of government policy by time series analysis in the period of enforcement of the reflux tax, from 2015 to 2017.

• Journal of the Korea Convergence Society
• /
• v.10 no.9
• /
• pp.161-166
• /
• 2019

This study examines the flows near the different side mirrors by analyzing the flow due to air resistance at A, B and C models of automotive side mirrors. Model A is a square-shaped side-mirror. Model B is a triangular side-mirror and model C is an oval-shaped side-mirror. The air resistance of the side-mirror while driving is reduced and the automotive power can be reduced by changing the design of automotive side-mirror. As analysis result, as the pressure of air resistance against side mirror becomes larger, it can be seen that the air flow rate becomes great. Therefore, it can be estimated that the smaller the pressure of air resistance, the smaller the flow rate and the better the air flow. Therefore, it can be acknowledged that model B is the best model. As the design data of the automotive side mirror obtained on the basis of this study result are utilized, the esthetic sense can be shown while driving a car at real life.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.4
• /
• pp.53-62
• /
• 2006

Two-dimensional steady flowfields generated by transverse injection jets into a supersonic mainstream are numerically simulated. Fine-scale turbulence effects are represented by a k-${\omega}$ SST two-equation closure model which includes $y^+$ effects on the turbulence model. Solution convergence is evaluated by using Grid Convergence Index(GCI), a measure of uncertainty of the grid convergence. Comparison is made with experimental data and other turbulence models in term of surface static pressure distributions, the length of the upstream separation region, and the penetration height. Results indicate that the k-${\omega}$ SST model correctly predicts the mean surface pressure distribution and the upstream separation length for low static pressure ratios. However, the numerical predictions become less consistent with experimental results as the static pressure ratio increases. All these results are taken within 1% error band of grid convergence.

• Journal of Digital Convergence
• /
• v.13 no.2
• /
• pp.333-341
• /
• 2015

The purpose of this research is to study possibilities of animation education in art centered STEAM. We suggest not the animation education contents model of a functional skills of the past but the one for STEAM. In order to suggest the role of animation education, we should first study the convergence values of animation. And this study make a curriculum of animation for the creativity by convergence thinking. Finally, this study build the animation class model applied in STEAM by Fogarty's integrated education method. By studying, we can get the result that animation has the convergence characteristics and it has the effective role of art centered STEAM education for the convergence thinking. Also, it is expected that the results of this study will contribute to education and training of convergence of the global talent with creativity in STEAM.