• Journal of Digital Contents Society
• /
• v.19 no.6
• /
• pp.1161-1167
• /
• 2018

The development of machine learning has enabled machines to perform delicate tasks that only humans could do, and thus many companies have introduced machine learning based translators. Existing translators have good performances but they have problems in number translation. The translators often mistranslate numbers when the input sentence includes a large number. Furthermore, the output sentence structure completely changes even if only one number in the input sentence changes. In this paper, first, we optimized a neural machine translation model architecture that uses bidirectional RNN, LSTM, and the attention mechanism through data cleansing and changing the dictionary size. Then, we implemented a number-processing algorithm specialized in number translation and applied it to the neural machine translation model to solve the problems above. The paper includes the data cleansing method, an optimal dictionary size and the number-processing algorithm, as well as experiment results for translation performance based on the BLEU score.

• Journal of Institute of Control, Robotics and Systems
• /
• v.19 no.6
• /
• pp.527-533
• /
• 2013

BCI (Brain-Computer Interface) is a system that transforms a subject's brain signal related to their intention into a control signal by classifying EEG (electroencephalograph) signals obtained during the imagination of movement of a subject's limbs. The BCI system allows us to control machines such as robot arms or wheelchairs only by imaging limbs. With the exact same experiment environment, activated brain regions of each subjects are totally different. In that case, a simple approach is to use as many channels as possible when measuring brain signals. However the problem is that using many channels also causes other problems. When applying a CSP (Common Spatial Pattern), which is an EEG extraction method, many channels cause an overfitting problem, and in addition there is difficulty using this technique for medical analysis. To overcome these problems, we suggest an optimal channel selection method using a BPSO (Binary Particle Swarm Optimization), BPSO with channel impact factor, and GA. This paper examined optimal selected channels among all channels using three optimization methods and compared the classification accuracy and the number of selected channels between BPSO, BPSO with channel impact factor, and GA by SVM (Support Vector Machine). The result showed that BPSO with channel impact factor selected 2 fewer channels and even improved accuracy by 10.17~11.34% compared with BPSO and GA.

• Journal of the Korean Society for Railway
• /
• v.18 no.4
• /
• pp.335-343
• /
• 2015

The role of the turnouts installed on the points of railway branches and intersections is to safely switch the directions of trains. In case of abnormal forces applied to the turnouts due to extreme operating conditions, it is difficult in the current system to conduct accurate diagnosis and measuring. Moreover, the existing measurement devices for turnout switching power are mostly based on foreign technologies, which provide only limited information on the switching power; this information is insufficient for effective monitoring and maintenance. In this paper, through an analysis of the characteristics of the switching and locking function, the problems in turnout power measurements of current electrical point machines in Korea are examined. And based on the results, a new sensor-based method for measuring switching power is proposed.

• Journal of Power Electronics
• /
• v.13 no.3
• /
• pp.369-380
• /
• 2013

Direct torque control based on space vector modulation (SVM-DTC) protects the DTC transient merits. Furthermore, it creates better quality steady-state performance in a wide speed range. The modified method of DTC using SVM improves the electrical magnitudes of asynchronous machines, such as minimizing the stator current distortions, the stator flux with electromagnetic torque without ripple, the fast response of the rotor speed, and the constant switching frequency. In this paper, the proposed method is based on two new control strategies for direct torque control with space vector modulation. First, fuzzy logic control is used instead of the PI torque and a PI flux controller to minimizing the torque error and to achieve a constant switching frequency. The voltages in the direct and quadratic reference frame ($V_d$, $V_q$) are achieved by fuzzy logic control. In this scheme, the switching capability of the inverter is fully utilized, which improves the system performance. Second, the close loop of stator flux estimation based on the voltage model and a low pass filter is used to counteract the drawbacks in the open loop of the stator flux such as the problems saturation and dc drift. The response of this new control strategy is compared with DTC-SVM. The experimental and simulation results demonstrate that the proposed control topology outperforms the conventional DTC-SVM in terms of system robustness and eliminating the bad outcome of dc-offset.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.11
• /
• pp.2131-2136
• /
• 2016

Before Industry 4.0, a purpose of manufacturing processes is simply overcoming the weakness of product line. However, Industry 4.0 should determine its method of work of facilities' own accord. Manufacturing processes such as FMS, AMS, MAS have been developed in these procedures, but there are possibilities of deadlock that can be happened for the reason of machine, agent or resource sharing. There were many studies to solve this problem. However, studies about prevention of deadlock happened in the situation that manufacturing processes are reconfigured dinamically because of human's intervention are incomplete. This paper proposes the model of the situation reconfigured dinamically and the method to prevent the deadlock in this situation. The proposed method adjusts production and manufacturing time by broadcasting time error messages to machines around when the manufacturing time of machine intervened by human is changed. Through this, it is anticipated to prevent the deadlock if conditions of the product line are changed.

• Journal of IKEEE
• /
• v.23 no.2
• /
• pp.543-551
• /
• 2019

Permanent Magnet Synchronous Motors(PMSMs) have a wider range of applications due to their high output density and high efficiency. PMSMs are used not only in high-power density, high-performance motor-driven systems such as vehicle and robots, but also in systems where cost-cutting is very important, such as washing machines, air conditioners and refrigerators. To reduce costs, position sensorless control is required, which is generally difficult to be used under conditions of starting the motor. Thus, the I-F speed control that rotates the current vector at any speed in the starting procedure should be used at first, and then the sensorless speed control could be applied after PMSM rotates above a certain speed. Speed control performance in I-F speed control and sensorless speed control is very important. And more speed control performance should be maintained even in the transient in which the two control techniques are changed. In this paper, the speed controller transition method from I-F speed control to sensorless speed control of permanent magnet synchronous motor is proposed. Experiments were carried out on the washing machine drive system to verify the performance of the proposed technique.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.3
• /
• pp.563-570
• /
• 2018

This paper proposes an efficient dissolved gas analysis(DGA) and classification method of an oil-filled transformer using machine learning algorithms to solve problems inherent in IEC 60599. In IEC 60599, a certain diagnosis criteria do not exist, and duplication area is existed. Thus, it is difficult to make a decision without any experts since the IEC 60599 standard can not support analysis and classification of gas date of a power transformer in that criteria. To address these issue. we propose a dissolved gas analysis(DGA) and classification method using a machine learning algorithm. We evaluate the performance of the proposed method using support vector machines with dissolved gas dataset extracted from a power transformer in the real industry. To validate the performance of the proposed method, we compares the proposed method with the IEC 60599 standard. Experimental results show that the proposed method outperforms the IEC 60599 in the classification accuracy.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.3
• /
• pp.571-578
• /
• 2018

The deteriorated pipline often causes the unexpected leakage and crack. Negligence and late maintenance leads the enormous damage for gas and water resource. This paper proposes early fault detection and diagnosis algorithm for pipeline using acoustic emission (AE) signals. Early fault detection method for pipeline compares the frequency amplitude of the spectrum to that of the spectrum in normal condition. Larger amplitude of the spectrum indicates abnormal condition. Early fault diagnosis algorithm uses support vector machines (SVM), which is trained for normal and abnormal conditions to diagnose the measured AE signal from the target pipeline. In the experiment, a pipeline testbed is constructed similarly to real industrial pipeline. Normal, 5mm cracked, 10mm holed pipelines are installed and tested in this study. The proposed fault detection and diagnosis technique is validated as an efficient approach to detect early faulty condition of pipeline.

• Journal of Power Electronics
• /
• v.15 no.5
• /
• pp.1244-1255
• /
• 2015

This paper presents a new multi-machine robust control based on an electric differential system for electric vehicle (EV) applications which is composed of four in-wheel permanent magnet synchronous motors. It is based on a new master-slave direct torque control (DTC) algorithm, which is used for the control of bi-machine traction systems based on a speed model reference adaptive system observer. The use of an electric differential in the design of a new EV constitutes a technological breakthrough. A classical system with a multi-inverter and a multi-machine comprises a three-phase inverter for each machine to be controlled. Another approach consists of only one three-phase inverter for several permanent magnet synchronous machines. The control of multi-machine single-inverter systems is the subject of this study. Several methods have been proposed for the control of multi-machine single-inverter systems. In this study, a new master-slave based DTC strategy is developed to generate an electric differential system. The entire system is simulated by Matlab/Simulink. The simulation results show the effectiveness of the new multi-machine robust control based on an electric differential system for use in EV applications.

• Proceedings of the KIEE Conference
• /
• 2006.04b
• /
• pp.226-228
• /
• 2006

This paper presents the two lossless auxiliary inducors-assisted voltage source type half bridge(single ended push pull:SEPP) series resonant high frequency inverter for induction heated fixing roller in copy and printing machines. The simple high-frequency inverter treated here can completely achieve stable zero current soft switching (ZCS) commutation forwide its output power regulation ranges and load variations under constant high frequency pulse density modulation (PDM) scheme. Its transient and steady state operatprinciple is originally described and discussed for a constant high-frequency PDM control strategy under a stable ZCS operation commutation, together with its output effective power regulation charactertics-based on the high frequency PDM strategy. The experimenoperating performances of this voltage source SEPP ZCS-PDM series resonant high frequency inverter using IGBTs are illustrated as compared with computer simulation results and experimenones. Its power losses analysis and actual efficiency are evaluated and discussed on the basis of simulation and experimental results. The feasible effectiveness of this high frequency inverter appliimplemented here is proved from the practical point of view.