• The Journal of the Korea institute of electronic communication sciences
• /
• v.12 no.6
• /
• pp.1065-1070
• /
• 2017

Magnetic gears are magnetically coupled to the input side and the output side of the rotary machine to transmit power without mechanical contact. The magnetic gear consists of an inner rotor, an outer rotor and pole pieces. Torque ripple occurs due to the difference in reluctance between the two rotors and the pole pieces during power transmission. Torque ripple is a cause of the noise and vibration of the rotary machine, so it is necessary to minimize it. In this paper, we propose a shape that cuts the corner of the pole piece and apply a fillet to reduce torque ripple. We used a two-dimensional finite element analysis method to compare and analyze the torque ripple of the magnetic gears according to the change of the fillet parameters and to find the pole piece shape with excellent torque ripple.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.6
• /
• pp.749-754
• /
• 2009

Two gait generation methods using GP(genetic programming) and CPG(Central Pattern Generator) are compared to develop a fast locomotion for quadruped robot. GP based technique is an effective way to generate few joint trajectories instead of the locus of paw positions and lots of stance parameters. The CPGs are neural circuits that generate oscillatory output from a input coming from the brain. Optimization for two proposed methods are executed and analysed using Webots simulation for the quadruped robot which is built by Bioloid. Furthermore, simulation results for two proposed methods are experimented in real quadruped robot and performances and motion features of GP and CPG based methods are investigated.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.19 no.3
• /
• pp.95-105
• /
• 2009

Spam call is one of the main security threat in VoIP services. In this paper, we have designed simulation model for performance evaluation of VoIP spam defense mechanism. The simulation model has functions for performance evaluation such as calls generation and input/output comparison. Four representative caller models have been developed for performance evaluation and each model has its own characteristics as statistical parameters. The target mechanism of performance evaluation is SPIT(Spam over Internet Telephony) level decision algorithm, and we have derived SPIT levels of caller models. The performance evaluation model is designed using the DEVS formalism and DEVSJAVA$^{TM}$ is exploited for development and execution of simulation models.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.3
• /
• pp.32-43
• /
• 2002

In this paper, we propose a pattern classification system for digital signal which is based on neural networks. The proposed system consists of two models of neural network. The first part is a wavelet neural network whose role is a feature extraction. For this part, we compare existing models of wavelet networks and propose a new model for feature extraction. The other part is a wavelet network for pattern classification. We modify the structure of previous wavelet network for pattern classification and propose a learning method. The inputs of the pattern classification wavelet network is connection weights, dilation and translation parameters in hidden nodes of feature extraction network. And the output is a class of the signal which is input of feature extraction network. The proposed system is, applied to classification of EEG signal based on frequency.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.39 no.7
• /
• pp.26-36
• /
• 2002

Mach-Zehnder type traveling-wave optical modulators with 3-section and 5-section phase reversal were designed and fabricated on z-cut $LiNbO_3$ substrates. Optical waveguides were designed by means of the FDM(Finite Difference Method). Design of CPW traveling-wave electrodes were performed by the SOR(Successive Over Relaxation) in the active region and by the CMM(Conformal Mapping) in the input/output section. The optical response R(${\omega}$ calculated based on the measured S-parameters showed the bandwidth of 15GHz centered at 25GHz for the 3-section, and the bandwidth of 22GHz at 45GHz for the 5-section.

• Smart Structures and Systems
• /
• v.22 no.4
• /
• pp.413-424
• /
• 2018

The factors affecting the shear strength of the angle shear connectors in the steel-concrete composite beams can play an important role to estimate the efficacy of a composite beam. Therefore, the current study has aimed to verify the output of shear capacity of angle shear connector according to the input provided by Support Vector Machine (SVM) coupled with Firefly Algorithm (FFA). SVM parameters have been optimized through the use of FFA, while genetic programming (GP) and artificial neural networks (ANN) have been applied to estimate and predict the SVM-FFA models' results. Following these results, GP and ANN have been applied to develop the prediction accuracy and generalization capability of SVM-FFA. Therefore, SVM-FFA could be performed as a novel model with predictive strategy in the shear capacity estimation of angle shear connectors. According to the results, the Firefly algorithm has produced a generalized performance and be learnt faster than the conventional learning algorithms.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.41 no.4
• /
• pp.321-327
• /
• 2017

In this study, to investigate the shape of the back bead as a weld quality parameter and to select the optimal condition of the root-pass TIG welding of a horizontal butt-joint, an experimental design and the response surface method (RSM) have been employed. Three parameters are used as input variables, which include the base current, peak current, and welding speed. The back bead width is selected as an output variable representing the weld quality, the target value of the width is 5.4 mm. Conducting the experiments according to the Box-Behnken experimental design, a $2^{nd}$ regression model for the back bead width was made, and the validation of the model was confirmed by using the F-test. The desirability function was designed through the nominal-the-best formula for the appropriate back bead width. Finally, the following optimal condition for welding was selected using the RSM: base current of 0.9204, peak current of 0.8676, and welding speed of 0.3776 in coded values. For verification, a test welding process under the optimal condition was executed and the result showed the back bead width of 5.38 mm that matched the target value well.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.4
• /
• pp.115-119
• /
• 2023

The natural language processing models used in current artificial intelligence are huge, causing various difficulties in processing and analyzing data in real time. In order to solve these difficulties, we proposed a method to improve the efficiency of processing by using less memory and checked the performance of the proposed model. The technique applied in this paper to evaluate the performance of the proposed model is to divide the large corpus by adjusting the number of attention heads and embedding size of the BERT[1] model to be small, and the results are calculated by averaging the output values of each forward. In this process, a random offset was assigned to the sentences at every epoch to provide diversity in the input data. The model was then fine-tuned for classification. We found that the split processing model was about 12% less accurate than the unsplit model, but the number of parameters in the model was reduced by 56%.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.15 no.3
• /
• pp.257-262
• /
• 2009

The unified power flow controller(UPFC) is one of the most effective devices in the FACTS family. This paper concerns about a filtering technique for reducing the computer calculation to determine the optimal location of UPFC in a power system. The sensitivities of the power generation cost for UPFC control parameters are evaluated. This technique requires that only one optimal power flow is run to get UPFC sensitivities for all possible transmission lines. To find out a optimal locating of a single UPFC in power system, an ideal transformer model which consists of a complex turns ratio and a variable shunt admittance was used. In this model, the UPFC control variables do not depend on UPFC input and output currents and voltages. The sensitivity method was tested on a 5-bus system derived from the IEEE 14-bus system and IEEE 14-bus system to establish its effectiveness.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.46 no.2
• /
• pp.1-8
• /
• 2009

We designed an ADC that operated upto 500Msamples/sec based on proposed R-string folding block as well as second folding block. The upper four bits are processed in parallel by the R-string folding block while the lower four bits are processed in pipeline structured second folding block to supply digital output. To verify the circuit performance, we conducted HSPICE simulation and the average power consumption was only 1.34mW even when the circuit was running at its maximum sampling frequency. We further measured noise immunity by applying linear ramp signal to the input. The DNL was between -0.56*LSB and 0.49*LSB and the INL was between -0.93*LSB and 0.72*LSB. We used 0.35 microns MOSIS device parameters for this work.