• 한국산업정보학회논문지
• /
• 제28권4호
• /
• pp.21-34
• /
• 2023

연합학습은 동기식 연합학습과 비동기식 연합학습으로 구분된다. 그 중에서 비동기식 연합학습은 동기식 연합학습 보다 시간적인 이득이 있으나 좋은 모델 성능을 얻기 위한 도전 과제가 남아있다. 특히 non-IID 학습 데이터셋에서 성능열화 방지, 적절한 클라이언트 선택 및 오래된 그래디언트 정보 관리는 모델 성능 개선에 있어 중요하다. 본 논문에서는 K-비동기식 연합학습을 다루고 있으며 non-IID 데이터셋을 통해 학습한다. 또한 기존 방식이 선택할 클라이언트 수에 있어서 정적인 K개를 사용한 것과 달리 동적으로 K 값을 조절하는 알고리즘을 제안하여 학습 시간을 줄일 수 있었다. 추가적으로, 오래된 그래디언트를 다루는 방식을 활용해 모델 성능 개선을 이루었음을 보여준다. 마지막으로 강한 모델 안정성을 얻기 위해 모델 성능을 평가하는 방식을 활용하였다. 실험 결과를 통해 전체 알고리즘을 활용했을 때 학습 시간 단축, 모델 정확도 향상, 모델 안정성 향상의 이득을 얻을 수 있음을 보여준다.

• 대한토목학회논문집
• /
• 제36권4호
• /
• pp.705-713
• /
• 2016

연약지반 공사는 설계단계에서 고려하지 못한 변수들이 시공단계에서 발생되어 구조물의 안정성과 설계변경에 따른 공사비 증가의 원인이 된다. 본 연구는 충청북도 오창지역에 조성되는 산업단지 진입도로 조성구간을 연구대상으로 하며, 연약지반 처리공법으로 암버럭을 이용한 치환공법 및 사용하중을 고려한 과재성토를 연약지반 처리대책으로 선정, 구조물의 안정성 및 공기 단축으로 공사비 절감효과 분석에 목적이 있다. 연구구간의 연약지반 두께는 사질토 및 점성토 연약지반으로 2.4~5.5m로 규모가 작으나 지표하 1.5~5.9m의 지층사이에 불균등하게 분포하며, 침하발생은 크지 않으나 장래 부등침하와 단차 발생이 예상되어 연약지반 처리후 현장 계측결과를 분석하여 장래 침하거동을 예측하였다. 또한, 치환재료는 지역특성 및 경제성을 고려하여 양질토사를 암버럭으로 치환하여 안정성 증가, 공기단축 등으로 공사비 29%를 절감하였다. 본 연구를 바탕으로 향후 제체내 간극수압 소산, 지하수위 변동 등의 예측과 실제 계측된 연약지반의 압밀과의 관계 등에 대한 추가적인 연구를 수행할 경우, 다양한 현장조건에서 연약지반 거동을 정확히 예측할 수 있을 것으로 기대된다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1996년도 하계학술대회 논문집 A
• /
• pp.559-563
• /
• 1996

Total 215 magnet power supplies are under operation in Pohang Light Source(PLS). Current, voltage and peak power handling capabilities of DC power supply are in the range of 21 to 643V, 45 to 850A and 0.94 to 531KW, respectively. The DC power supplies can be categoriezed as uni-polar and bipolar power. Typical required stability and ripple of uni-polar power supplies are ${\pm}0.005%$ and ${\pm}0.05%$, respectively. The bipolar power supplies should be maintained within ${\pm}0.05%$ stability and ${\pm}0.5%$ ripple. Precise measurement results show that all power supplies meet or exceed the required specifications. The long term operation reliability to be sufficient for a stable operation of the Pohang Light Source.

• 제어로봇시스템학회논문지
• /
• 제7권8호
• /
• pp.664-674
• /
• 2001

In this paper, as a new category of fuzzy-neural networks architecture, we propose Fuzzy Polynomial Neural Networks (FPNN) and discuss a comprehensive design methodology related to its architecture. FPNN dwells on the ideas of fuzzy rule-based computing and neural networks. The FPNN architecture consists of layers with activation nodes based on fuzzy inference rules. Here each activation node is presented as Fuzzy Polynomial Neuron(FPN). The conclusion part of the rules, especially the regression polynomial, uses several types of high-order polynomials such as linear, quadratic and modified quadratic. As the premise part of the rules, both triangular and Gaussian-like membership functions are studied. It is worth stressing that the number of the layers and the nods in each layer of the FPNN are not predetermined, unlike in the case of the popular multilayer perceptron structure, but these are generated in a dynamic manner. With the aid of two representative time series process data, a detailed design procedure is discussed, and the stability is introduced as a measure of stability of the model for the comparative analysis of various architectures.

• Journal of Electrical Engineering and Technology
• /
• 제6권4호
• /
• pp.484-492
• /
• 2011

Power system analyses, which involve the handling of massive data volumes, necessitate the use of effective visualization methods to facilitate analysis and assist the user in obtaining a clear understanding of the present state of the system. This paper introduces an interface that compensates for the limitations of the visualization modules of dynamic security assessment tools, such as PSS/e and TSAT, for power system variables including generator rotor angle and frequency. The compensation is made possible through the automatic provision of dynamic simulation data in visualized and tabular form for better data intuition, thereby considerably reducing the redundant manual operation and time required for data analysis. The interface also determines whether the generators are stable through a generator instability algorithm that scans simulation data and checks for an increase in swing or divergence. The proposed visualization methods are applied to the dynamic simulation results for contingencies in the Korean Electric Power Corporation system, and have been tested by power system researchers to verify the effectiveness of the data visualization interface.

• 한국항해학회지
• /
• 제24권5호
• /
• pp.373-383
• /
• 2000

The ship’s safe mooring stability is a principles for the safe cargo handling works at the mooring berth. Today numerous standards, guidelines and recommendations concerning mooring practices, fittings and equipments exist throughout the worldwide maritime industries. In recently, the mooring facilities were constructed as dolphin types at the open sea area apart far from shoreside instead of enclosed coastline area in accordance with increasing ship’s size and for preventing environmental pollution. Therefore the exciting wave condition must be considered as a basic environmental criteria with the wind force and current force for all of the mooring ships at the sea berth facilities. In this study, this added wave force as one of the environmental external forces by using the theoretical formula was applied to the LNG ship in Pyeongtaeg harbor needed the special mooring stability of the sea berth. Through this research, it can be confirmed that wave force is the very important factor in the mooring force and the strength of wave force works much more in the full laden condition than in the lightship condition. And also the wave force changes to non-linear states according to the wave frequency and wave length. In addition, the maximum limit criteria of environmental force of prohibiting the entering ship on the berth and loading works controlled by the port authority concerned of Pyeongtaeg port fully satisfies the condition of the mooring limit force recommended by OCIMF that the safe permitted force of the mooring line have to be within 55 % of MBL.

• International Journal of Aeronautical and Space Sciences
• /
• 제8권1호
• /
• pp.32-45
• /
• 2007

The T-50 advanced supersonic jet trainer employs the Relaxed Static Stability (RSS) concept to improve the aerodynamic performance while the flight control system stabilizes the unstable aircraft and provides adequate handling qualities. The T-50 flight control laws employ a proportional-plus-integral type controller based on a dynamic inversion method in longitudinal axis and a proportional type controller based on a blended roll system with simple roll rate feedback and beta-betadot feedback system. These control laws are verified by flight tests with various maneuver set flight envelopes and the control laws are updated to resolve flight test issues. This paper describes several concepts of flight control laws used in T-50 to resolve those flight test issues. Control laws for solving the roll-off problem during pitch maneuver in asymmetric loading configurations, improving the departure resistance in negative angle of attack conditions and enhancing the fine tracking performance in air-to-air tracking maneuvers are described with flight test data.

• 제어로봇시스템학회논문지
• /
• 제14권3호
• /
• pp.307-317
• /
• 2008

Relaxed static stability(RSS) concept has been applied to improve aerodynamic performance of modem version supersonic jet fighter aircraft. Therefore, flight control system are necessary to stabilize an unstable aircraft and provides adequate handling qualities. Also, flight control systems of modem version aircraft employ a safety system to support emergency situations such as a pilot unknown attitude flight conditions of an aircraft in night flight-testing. This situation is dangerous because the aircraft can lose if the pilot not take recognizance of situation. Therefore, automatic recovery system is necessary. The system called the "Pilot Activated Recovery System" or PARS, provided a pilot initiated automatic maneuver capable of an aircraft recoveries in situations of unusual attitudes. This paper addresses the concept of PARS and designed using nonlinear control law design process based on model of supersonic jet trainer. And, this control law is verified by nonlinear analysis and real-time pilot evaluation using in-house software. The result of evaluation reveals that the PARS support recovery of an aircraft unusual attitude and improve a safety of an aircraft.

• 로봇학회논문지
• /
• 제19권1호
• /
• pp.31-38
• /
• 2024

A mobile manipulator is capable of handling a wide range of workspaces by overcoming the limitations of mobility inherent in existing fixed-base manipulators. To simulate the mobile manipulator, two contact operations should be considered in the physics engines. One of these operations is the grasp stability between the gripper and the object, while the other involves the contact between the wheels of the mobile robot and the ground during driving. However, it is still difficult to choose an appropriate physics engine for simulating these contact operations of the mobile manipulator. In this paper, the performance of physics engines for simulating the mobile manipulator is evaluated. Firstly, the grasp stability of the physics engine is quantitatively evaluated based on the contact force discontinuity. Secondly, when the mobile robot is controlled by open or closed-loop control methods, differences in the path taken by the mobile robot depending on the physics engine are analyzed. To assess the performance of robot simulation, three dynamic simulators-MuJoCo, CoppeliaSim, and IsaacSim-are used along with five physics engines: MuJoCo, Newton, ODE, Bullet, and PhysX.

• 항공우주시스템공학회지
• /
• 제16권1호
• /
• pp.17-27
• /
• 2022

실제와 모델간의 다양한 불확실성(uncertainties)에 대한 항공기의 강건성 확보를 위해 비행제어시스템은 일정 수준 이상의 안정성 여유(stability margin) 감항기준을 만족할 수 있도록 설계되어야 한다. 미 국방 규격인 MIL-F-9490D와 감항인증 규격인 MIL-HDBK-516B에는 비행제어시스템이 만족해야 하는 이득여유(gain margin)와 위상여유(phase margin) 기준을 제시하고 있다. 그러나 항공기의 개발 단계별 설계 성숙도를 고려하지 않고 전 개발단계에서 동일한 안정성 여유 기준이 적용되면서, 궁극적으로는 항공기 운용영역이 과도하게 제한되는 문제가 발생한다. 본 논문에서는 조종성과 안정성 여유의 관계를 제시하고 개발 단계와 입증 방법에 따라 안정성 여유 기준을 재정립하는 방안을 제시한다. 연구의 결과는 현재 개발되고 있는 항공기뿐만 아니라 향후에 전개될 무인 비행체 개발 시에 적용하여 보다 유연하고 효과적인 방법으로 안정성 여유 기준을 입증하는데 기여하리라 판단한다.