• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 하계학술대회 논문집 B
• /
• pp.552-554
• /
• 1998

In target tracking problems the fixed gain Kalman filter is primarily used to predict a target state vector. This filter, however, has a poor precision for maneuvering targets while it has a good performance for non-maneuvering targets. To overcome the problem this paper proposes the system which estimates the acceleration with neural networks using the input estimation technique. The ability to efficiently fuse information of different forms is one of the major capabilities of trained multi-layer neural networks. The primary motivation for employing neural networks in these applications comes from the efficiency with which more features can be utilized as inputs for estimating target maneuvers. The parallel processing capability of a properly trained neural network can permit fast processing of features to yield correct acceleration estimates. The features used as inputs can be extracted from the combinations of innovation data and heading changes, and for this we set the two dimensional model. The properly trained neural network system outputs the acceleration estimates and compensates for the primary Kalman filter. Finally the proposed system shows the optimum performance.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 제37회 하계학술대회 논문집 B
• /
• pp.675-676
• /
• 2006

This paper shows the temperature rise of the high voltage GIS bus bar. The temperature rise in GIS bus bar is due to Joule's losses in the conductor and the induced eddy current in the tank. The power losses of a bus bar calculated from the magnetic field analysis are used as the input data for the thermal analysis to predict the temperature. The required analysis is a couple-field Multiphysics that accounts for the interactions between three-dimensional AC harmonic magnetic and fluid fields. The heat transfer calculation using the fluid analysis is done by considering the natural convection and the radiation from the tank to the atmosphere. Consequently, because temperature distributions by couple-field Multiphysics (coupled magnetic-fluid) have good agreement with results of temperature rise test, the proposed couple-field Multiphysics technique is likely to be used in a conduction design of the single-pole and three pole-encapsulated bus bar in CIS..

• Structural Engineering and Mechanics
• /
• 제37권2호
• /
• pp.131-147
• /
• 2011

The logarithmic decrement method has been long used to estimate damping ratios in systems with only one modal component such as linear single degree of freedom (SDOF) mechanical systems. This paper presents an application of a methodology that uses joint time-frequency distribution (JTFD) as input, instead of the raw signal, to systems with several vibration modes. A most important feature of the present approach is that it can be applied to a system with time-varying damping ratio. Initially the precision and robustness of the method is determined using a synthetic model with multiple harmonic components, one of them displaying a time-varying damping ratio, subsequently the results obtained from experiments with a reduced model are presented. A comparison is made between the results obtained with this methodology and those using the classical technique of Least Squares Complex Exponential Method (LSCE) in order to highlight the advantages of the former, such as, good precision, robustness and excellent performance in extreme cases, e.g., when very low frequency components and time varying damping ratio are present.

• 전기학회논문지P
• /
• 제61권4호
• /
• pp.186-191
• /
• 2012

DC Bus Electrolytic capacitors have been widely used in power conversion system because they can achieve high capacitance and voltage ratings with volumetric efficiency and low cost. This type of capacitors have been traditionally used for filtering, voltage smoothing, by-pass and other many applications in power conversion circuits requiring a cost effective and volumetric efficiency components. Unfortunately, electrolytic capacitors are some of the weakest components in power electronic converter. Many papers have proposed different methods or algorithms to determinate the ESR and/or capacitance C for fault diagnosis of the electrolytic capacitor. However, both ESR and C vary with frequency and temperature. Accurate knowledge of both values at the capacitors operating conditions is essential to achieve the best reference data of fault judgement. According to parameter analysis, the capacitance increases with temperature and the ESR decreases. Higher frequencies make the ESR and C to decrease. Analysis results show that the proposed electrolytic capacitor parameter estimation technique can be applied to reference signal of capacitor diagnosis systems successfully.

• 대한전자공학회논문지TC
• /
• 제45권12호
• /
• pp.26-33
• /
• 2008

신호의 검파는 무선통신시스템에서 매우 중요한 문제이며 최근에는 다중입력다중출력(MIMO) 통신을 위한 검파 알고리즘에 대한 연구가 많이 진행되었다. 그러나 이러한 연구의 많은 부분이 주파수 비선택적 채널을 가정하였거나 주파수 선택적 채널을 가정하더라도 수신단에서의 채널상태정보는 완전하다는 가정 하에 연구가 수행되었다. 따라서, 본 연구에서는 주파수 선택적 채널에서 불완전한 채널정보를 사용할 때 몇 가지 MIMO 검파 알고리즘에 대해 얻을 수 있는 오류율 성능을 비교해 보고자 한다.

• 한국통신학회논문지
• /
• 제34권5B호
• /
• pp.443-448
• /
• 2009

본 논문에서는 에너지가 제한적인 무선 센서네트워크에서 MIMO 기반 협력통신을 가능하게 하는 훈련심볼의 오버헤드와 데이터 수집노드(DGN)의 속도의 임계값을 제안한다. 두개의 송수신 안테나가 있는 경우에 대하여 에너지 효율과 지연값에 대한 성능을 분석하였다. 센서로부터 데이터 수집노드까지 장거리 통신을 할 경우에 대하여 기존의 SISO 보다 에너지를 적게 소모하는 MIMO 기반 협력통신 무선 센서 네트워크의 속도와 훈련심볼의 오버헤드 임계값들을 구하였으며 이들의 상관관계도 분석하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1993년도 하계학술대회 논문집 A
• /
• pp.227-230
• /
• 1993

Recently, the development of novel control methodology enables us to improve the performance of AC-machine drives by using pulse width modulation (PWM) technique. Usually, the dynamic characteristic of induction motor (IM) has been represented by the 5-th order nonlinear differential equation. This dynamics, however, can be reduced to 3-rd order dynamics by applying direct control of IM input current. This methodology concludes that it is much easier to control IM by means of the field-oriented methods employing the current controller. Therefore a precise current control is crucial to achieve a high control performance both in dynamic and steady state operations. This paper presents an adaptive fuzzy current controller with artificial neural network (ANN) for field-oriented controlled IM. This new control structure is able to adaptively minimize a current ripple while maintaining constant switching frequency. Especially the proposed controller employs neuro-computing philosophy as well as adaptive learning pattern recognizing principles with respect to variations of the system parameters. The proposed approach is applied to the IM drive system, and its performance is tested through various simulations. Simulation results show that the proposed system, compared among several known classical methods, has a superb performance.

• 로봇학회논문지
• /
• 제17권3호
• /
• pp.264-272
• /
• 2022

Human-robot cooperative tasks are increasingly required in our daily life with the development of robotics and artificial intelligence technology. Interactive reinforcement learning strategies suggest that robots learn task by receiving feedback from an experienced human trainer during a training process. However, most of the previous studies on Interactive reinforcement learning have required an extra feedback input device such as a mouse or keyboard in addition to robot itself, and the scenario where a robot can interactively learn a task with human have been also limited to virtual environment. To solve these limitations, this paper studies training strategies of robot that learn table balancing tasks interactively using deep reinforcement learning with human's facial expression feedback. In the proposed system, the robot learns a cooperative table balancing task using Deep Q-Network (DQN), which is a deep reinforcement learning technique, with human facial emotion expression feedback. As a result of the experiment, the proposed system achieved a high optimal policy convergence rate of up to 83.3% in training and successful assumption rate of up to 91.6% in testing, showing improved performance compared to the model without human facial expression feedback.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권8호
• /
• pp.2736-2750
• /
• 2022

As demands for the maritime communications increase, a variety of functions and information are required to exchange via elements of maritime systems, which leads communication traffic increases in maritime frequency bands, especially in VHF (Very High Frequency) band. Thus, effective resource management is crucial to the future maritime communication systems not only to the typical terrestrial communication systems. VHF data exchange system (VDES) enables to utilize more flexible configuration according to the communication condition. This paper focuses on the VDES communication system among VDES terminals such as shore stations, ship stations and aids to navigation (AtoN) to address efficient resource allocation. We propose a resource management method considering a MIMO (Multiple Input Multiple Output) technique in VDES, which has been widely used for modern terrestrial wireless networks but not for marine environments by scheduling the essential communication resources. We introduce the general channel model in marine environment and give two metrics, spectral and the energy efficiencies to examine our resource scheduling algorithm. Based on the simulation results and analysis, the proposed method provides a possibility to enhance spectral and energy efficiencies. Additionally, we present a trade-off relationship between spectral and energy efficiencies. Furthermore, we examine the resource efficiencies related to the imperfect channel estimation.

• Geomechanics and Engineering
• /
• 제37권3호
• /
• pp.253-262
• /
• 2024

Dynamic properties are pivotal in soil analysis, yet their experimental determination is hampered by complex methodologies and the need for costly equipment. This study aims to predict dynamic soil properties using static properties that are relatively easier to obtain, employing machine learning techniques. The static properties considered include soil cohesion, friction angle, water content, specific gravity, and compressional strength. In contrast, the dynamic properties of interest are the velocities of compressional and shear waves. Data for this study are sourced from 26 boreholes, as detailed in a geotechnical investigation report database, comprising a total of 130 data points. An importance analysis, grounded in the random forest algorithm, is conducted to evaluate the significance of each dynamic property. This analysis informs the prediction of dynamic properties, prioritizing those static properties identified as most influential. The efficacy of these predictions is quantified using the coefficient of determination, which indicated exceptionally high reliability, with values reaching 0.99 in both training and testing phases when all input properties are considered. The conventional method is used for predicting dynamic properties through Standard Penetration Test (SPT) and compared the outcomes with this technique. The error ratio has decreased by approximately 0.95, thereby validating its reliability. This research marks a significant advancement in the indirect estimation of the relationship between static and dynamic soil properties through the application of machine learning techniques.