• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 추계학술대회논문집
• /
• pp.97-100
• /
• 2007

This study deals with the sound insulation performance of the damping sheet, which is widely used in a train. The wall of train is assumed to be a plate made of steel and two damping sheets. In case of damping sheet only, the sound transmission loss follows the mass law. If a steel plate is attached to damping sheet, the transmission loss is found to be higher than single of only steel plate, as we can anticipated, about $3{\sim}5$ dB. This is very well known consequence that is because the density of area increases. However, the increase of the transmission loss is not higher than what we can expect by the mass law. That's because the steel is perfectly blocked from the transmission of the air ; There is no defect in that material.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제53권10호
• /
• pp.678-689
• /
• 2004

The major concern of networked control system is network uncertainties such as time delay and data loss. Because these uncertainties may degrade the performance of networked control system and destabilize the entire system. Therefore, the performance and the stability variation of networked control system due to network uncertainties must be considered first in designing networked control system. In particular, the stability analysis of networked control system is most important issue since time delay and data loss can make the overall systems unstable. In this paper, we present a new stability analysis method of networked control system with time delay and data loss, which is impossible in previous works. The proposed method can determine maximum time delay and allowable transmission rate that preserve stability performance of networked control system. The results of the simulation validate effectiveness of our stability analysis method.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2004년도 춘계학술대회논문집
• /
• pp.718-723
• /
• 2004

The insertion loss of a noise barrier comes from the effects of diffraction, transmission loss, absorption coefficient and attenuation by direct propagation. The noise level after the noise barrier, differs reatly from the diffraction on the upper part of the barrier. Maekawa, furze and Anderson presented a empirical formula for calculating the diffraction of a semi infinte screen shaped noise barrier. In this syudy, Noise reduction performance software was developed for the proper design and assessment of new plastic barrier . Predicted sound pressure level from using the software is compared with the site-measurement results to verify the noise reduction performance and feasibility of prediction software for insertion loss of noise barrier.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2000년도 춘계학술대회논문집B
• /
• pp.574-579
• /
• 2000

In this study. an off-design performance prediction program for centrifugal pumps is developed. To estimate the losses in an impeller flow passage, two-zone model and two-element in series(TEIS) model are used. At impeller exit. the mixing process occurs with an increase in entropy. In two-zone model. there are both primary zone and secondary zone for an isentropic core flow and an average of all non-isentropic streamtubes respectively. The level of the core flow diffusion in an impeller was calculated by using TEIS model. While internal losses in an impeller an automatically estimated by using the above models, some empirical correlations far estimating external losses. far example, disk friction loss, recirculation loss and leakage loss are used. In order to analyze the vaneless diffuser flow. the momentum equations for the radial and tangential directions are used and solved together with continuity and energy equations.

• Journal of Communications and Networks
• /
• 제5권1호
• /
• pp.82-89
• /
• 2003

In this paper, an efficient threshold-based filtering (TF) buffer management scheme is proposed. The TF is capable of minimizing the overall loss performance and improving the fairness of buffer usage in a shared buffer packet switch. The TF consists of two mechanisms. One mechanism is to classify the output ports as sctive or inactive by comparing their queue lengths with a dedicated buffer allocation factor. The other mechanism is to filter the arrival packets of inactive output ports when the total queue length exceeds a threshold value. A theoretical queuing model of TF is formulated and resolved for the overall packet loss probability. Computer simulations are used to compare the overall loss performance of TF, dynamic threshold (DT), static threshold (ST) and pushout (PO). We find that TF scheme is more robust against dynamic traffic variations than DT and ST. Also, although the over-all loss performance between TF and PO are close to each other, the implementation of TF is much simpler than the PO.

• 대한임베디드공학회논문지
• /
• 제6권3호
• /
• pp.149-156
• /
• 2011

Transmission delay and packet loss induced by a communication network can degrade the control performance and, even make the system unstable. This paper presents a method for compensating transmission delay and packet loss in a networked control system for unmanned underwater vehicle. The proposed method is based on Lagrange interpolation in order to satisfy the requirements of simplicity and model-independency. In this work, the lost/delayed data are estimated in real time by only using the past data without requiring any mathematical model of the controlled system. Consequently, the proposed method can be implemented independent of the controlled system, and also it can achieve fast and accurate compensation performance. The performance of the proposed technique is evaluated by numerical simulations with an unmanned underwater vehicle.

• 한국추진공학회:학술대회논문집
• /
• 한국추진공학회 2008년 영문 학술대회
• /
• pp.202-212
• /
• 2008

A three-dimensional computation was conducted to understand effects of the low Reynolds number on the loss characteristics in a transonic axial compressor, Rotor67. As a gas turbine becomes smaller in size and it is operated at high altitude, the operating condition frequently lies at low Reynolds number. It is generally known that wall boundary layers are thickened and a large separation occurs on the blade surface in axial turbomachinery as the Reynolds number decreases. In this study, it was found that the large viscosity did not affect on the bow shock at the leading edge but significantly did on the location and the intensity of the passage shock. The passage shock moved upstream towards leading edge and its intensity decreased at the low Reynolds number. This change had large effects on the performance as well as the internal flows such as the pressure distribution on the blade surface, tip leakage flow and separation. The total pressure rise and the adiabatic efficiency decreased about 3% individually at the same normalized mass flow rate at the low Reynolds number. In order to analyze this performance drop caused by the low Reynolds number, the total pressure loss was scrutinized through major loss categories such as profile loss, tip leakage loss, endwall loss and shock loss.

• 한국인터넷방송통신학회논문지
• /
• 제24권1호
• /
• pp.49-54
• /
• 2024

데이터 불균형은 분류 문제에서 흔히 마주치는 문제로, 데이터셋 내의 클래스간 샘플 수의 현저한 차이에서 기인한다. 이러한 데이터 불균형은 일반적으로 분류 모델에서 과적합, 과소적합, 성능 지표의 오해 등의 문제를 야기한다. 이를 해결하기 위한 방법으로는 Resampling, Augmentation, 규제 기법, 손실 함수 조정 등이 있다. 본 논문에서는 손실 함수 조정에 대해 다루며 특히, 불균형 문제를 가진 Multi-Class 블랙박스 동영상 데이터에서 여러 구성의 손실 함수(Cross Entropy, Balanced Cross Entropy, 두 가지 Focal Loss 설정: 𝛼 = 1 및 𝛼 = Balanced, Asymmetric Loss)의 성능을 I3D, R3D_18 모델을 활용하여 비교하였다.

• 한국통신학회논문지
• /
• 제33권7B호
• /
• pp.477-487
• /
• 2008

TCP에서의 혼잡제어는 패킷 손실이 발생하면 이를 네트워크의 혼잡상황으로 판단해서 전송률을 줄인다. 무선 네트워크에서는 채널 에러로 인해 패킷 손실이 발생하는데, 기존의 유선환경에서의 TCP는 이를 혼잡으로 인한 손실로 착각하여 성능을 떨어뜨리는 결과를 초래한다. 그러므로 유 무선 통합네트워크에서의 TCP 성능 저하를 막기 위해 혼잡손실과 무선손실을 구별하는 연구가 진행되고 있다. 기존의 무선 TCP에 대한 연구는 주로 패킷이 전달되는 시간의 변화를 통해 네트워크의 혼잡상황을 유추해서 패킷 손실 시 혼잡손실과 무선손실을 예측하지만, 패킷의 전송시간은 여러 가지 다른 요인에 영향을 받기 때문에 정확한 손실구분은 불가능하다. 그러므로 본 논문에서는 IEEE 802.11 MAC에서 정의하고 있는 MIB(Management Information Base)의 무선손실 정보를 이용하여 유선손실과 무선손실을 구별하는 알고리즘을 제안한다. MAC 계층의 MIB를 수집하여 사용하는 제안된 알고리즘과 패킷의 지연 시간을 이용하는 기존의 알고리즘을 시뮬레이션을 통하여 비교하고 분석한 결과 무선 채널에서의 에러율이 10%인 경우에, Spike 알고리즘에 비해 12%, mBiaz 알고리즘에 비해 32%의 성능 향상을 보였다.

• 한국산업융합학회 논문집
• /
• 제3권3호
• /
• pp.265-273
• /
• 2000

Basic simulation program for Vuilleumier cycle heat pump was developed that can use precise VMHP design and analysis. VMHP system was divided 11 sections in simulation. Simulation was used adiabatic model analysis and that considered with heat transfer performance for heat exchanger, regenerator loss, conduction loss, shuttle loss, pumping loss and pressure loss by flow friction. Specially, friction loss of connection pipe between heat compression side and heat pump side, leakage of rod seal and piston seal was considered in the analysis.