• Journal of Electrical Engineering and Technology
• /
• 제12권6호
• /
• pp.2219-2226
• /
• 2017

This paper deals with artificial neural network approach for automatic detection of severity level of stator winding fault in induction motor. The problem is faced through modelling and simulation of induction motor with inter coil shorting in stator winding. The sum of the absolute values of difference in the peak values of phase currents from each half cycle has been chosen as the main input to the classifier. Sample values from workspace of Simulink model, which are verified with experiment setup practically, have been imported to neural network architecture. Consideration of a single input extracted from time domain simplifies and advances the fault detection technique. The output of the feed forward back propagation neural network classifies the short circuit fault level of the stator winding.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1994년도 Proceedings of the Korea Automatic Control Conference, 9th (KACC) ; Taejeon, Korea; 17-20 Oct. 1994
• /
• pp.737-741
• /
• 1994

We show an application of a genetic algorithm to, control systems including neural networks. Genetic algorithms are getting more popular nowadays because of their simplicity and robustness. Genetic algorithms are global search techniques for optimization and many other problems. A feed-forward neural network which is widely used in control applications usually learns by error back propagation algorithm(EBP). But, when there exist certain constraints, EBP can not be applied. We apply a modified genetic algorithm to such a case. We show simulation examples of two cart-pole nonlinear systems: single pole and double pole.

• Structural Engineering and Mechanics
• /
• 제51권2호
• /
• pp.299-313
• /
• 2014

In the first part of this paper, the influences of some of crack parameters on natural frequencies of a cracked cantilever Functionally Graded Beam (FGB) are studied. A cantilever beam is modeled using Finite Element Method (FEM) and its natural frequencies are obtained for different conditions of cracks. Then effect of variation of depth and location of cracks on natural frequencies of FGB with single and multiple cracks are investigated. In the second part, two Multi-Layer Feed Forward (MLFF) Artificial Neural Networks (ANNs) are designed for prediction of FGB's Cracks' location and depth. Particle Swarm Optimization (PSO) and Back-Error Propagation (BEP) algorithms are applied for training ANNs. The accuracy of two training methods' results are investigated.

• 한국정보디스플레이학회:학술대회논문집
• /
• 한국정보디스플레이학회 2002년도 International Meeting on Information Display
• /
• pp.567-570
• /
• 2002

The LVDS (Low Voltage Differential Signaling) I/O cells, fully compatible with ANSI TIA/ EIA-644 LVDS standard, are designed using a 0.35${\mu}m$ standard CMOS technology. With a single 3V supply, the core cells operate at 1.34Gbps and power consumption of the output driver and the input receiver is 10. 5mW and 4.2mW, respectively. In the output driver, we employ the DCMFB (Dynamic Common-Mode FeedBack) circuit which can control the DC offset voltage of differential output signals. The SPICE simulation result of the proposed output driver shows that the variation of the DC offset voltage is 15.6% within a permissible range. In the input receiver, the proposed dual input stage with a positive feedback latch covers rail-to-rail input common-mode range and enables a high-speed, low-power operation. 5-channels of the proposed LVDS I/O pair can handle display data up to 8-bit gray scale and UXGA resolution.

• 천문학회보
• /
• 제37권2호
• /
• pp.219.2-219.2
• /
• 2012

We are developing Adaptive Optics (AO) system for 24 inch telescope at Seoul National University Observatory. It consists of the tip-tilt correction system and the residual wavefront error correction system with a deformable mirror and a wavefront sensor. We present the construction and performance measurements of the tip-tilt correction system. The tip-tilt component is the single largest contributor to wavefront error, especially for small telescope. The tip-tilt correction system consists of a quadrant photodiode, a tip-tilt mirror and a feed back loop. The collimated He-Ne laser beam is used for input light source and is artificially disturbed by air turbulence generated by a heat gun. Most of the turbulence is of low frequency less than 20 Hz, but extends to a few hundreds Hz. It is found that the closed loop system using proportional-integral-derivative (PID) control successfully corrects tip-tilt error at a rate as high as 300~400 Hz.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2018년도 학술발표회
• /
• pp.157-157
• /
• 2018

A reliable and accurate downscaling model which can provide climate change information, obtained from global climate models (GCMs), at finer resolution has been always of great interest to researchers. In order to achieve this model, linear methods widely have been studied in the past decades. However, nonlinear methods also can be potentially beneficial to solve downscaling problem. Therefore, this study explored the applicability of some nonlinear machine learning techniques such as neural network (NN), extreme learning machine (ELM), and ELM autoencoder (ELM-AE) as well as a linear method, least absolute shrinkage and selection operator (LASSO), to build a reliable temperature downscaling model. ELM is an efficient learning algorithm for generalized single layer feed-forward neural networks (SLFNs). Its excellent training speed and good generalization capability make ELM an efficient solution for SLFNs compared to traditional time-consuming learning methods like back propagation (BP). However, due to its shallow architecture, ELM may not capture all of nonlinear relationships between input features. To address this issue, ELM-AE was tested in the current study for temperature downscaling.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2001년도 춘계학술대회논문집E
• /
• pp.288-293
• /
• 2001

This paper presents the water-hammer effect due to the rapid opening and closing of isolation valve and thruster valve in the spacecraft propulsion system. The single propellant feed system was modeled to investigate the maximum peak pressure due to the water-hammer effect. The test parameters are tank supply pressure, shape and throat length of orifice and line length. Kerosene was used as the inert simulant propellant liquid instead of hydrazine. As downstream line length after isolation valve increased from 1.5 to 2.5m, the maximum line-filling water-hammer peak pressure decreased, but the average time interval between peak pressures increased. The maximum line-filling water-hammer peak pressure with orifice was lower than without orifice, and the maximum line-filling water-hammer peak pressure with orifice at the back of isolation valve was lower than with orifice in front of isolation valve. Without orifice, the maximum water-hammer peak pressure due to the rapid opening and closing of the thruster valve was about 126% of tank supply pressure. With orifice, it decreased. As orifice throat length increased, it decreased. The maximum water-hammer peak pressure due to the rapid closing of the thruster valve with converging-diverging orifice was lower than normal orifice. It was found that the orifice as a means of pressure drop was very effective to reduce the water hammer peak pressure at the thruster valve. The results of this study can be used for the design of spacecraft liquid propulsion feed system.

• 한국정보통신학회논문지
• /
• 제21권3호
• /
• pp.499-508
• /
• 2017

본 논문에서는 이중 대역 송신시스템에서 비선형 전력증폭기를 선형화하기 위한 새로운 디지털 전치왜곡 기법을 제안한다. 기존의 시스템에서는 자기 대역의 비선형뿐만 아니라 타 대역간의 비선형을 보상하기 위해 두 개의 피드백 경로가 요구되는데 이는 하드웨어의 복잡도와 비용을 증가시킨다. 반면 제안하는 전치왜곡 시스템은 한 개의 피드백 경로만을 사용한다. 따라서 기존 시스템에 비해 훨씬 간단한 구조를 가진다. 제안하는 기법은 전치왜곡 계수를 얻기 위해서 한 개의 피드백 경로를 공유하여 두 대역의 전력증폭기 특성을 먼저 추정하고 추정된 전력증폭기 특성으로부터 전치왜곡 계수를 얻는다. 컴퓨터 모의실험 결과에 따르면 제안된 방식은 두 개의 피드백 회로를 구성한 기존 방식과 대등한 선형화 성능을 보인다.

• Asian-Australasian Journal of Animal Sciences
• /
• 제13권6호
• /
• pp.802-810
• /
• 2000

A total of 120 finishing crossbred pigs ($Landrace{\times}Large$ $White{\times}Duroc$) with equal numbers of barrows and gilts weighing 58.5 kg body weight were used in a feeding trial, and 6 pigs (three of each sex) were used in a metabolic trial to investigate the effect of phase feeding. Finishing period was divided into two phases and 4 different diets were fed for those periods. Growth performance was not significantly different among treatments within the same sex. This result showed that 16% crude protein for early finishing period and 14% crude protein diet for late finishing period should be optimum. During the early finishing period, only feed intake was significantly different between sexes (p<0.01), but in late finishing period daily weight gain (p<0.001) and feed intake (p<0.01) of barrows were significantly higher than those of gilts. During the early finishing period, digestibilities of dry matter, protein and phosphorus were significantly higher in gilts than in barrows (p<0.05). However, there was no treatment effect within same sex during the early and late finishing period. During early finishing period, excretion of N of pigs fed 16% CP diet in early and 14% CP diet in late-finishing period was less than that of pigs fed 17% CP diet in early and 15% CP diet in late-finishing period (p<0.05), but the difference was not significant. During the late finishing period, N excretion with two phase feeding was reduced by 8.5% compared with single feeding. In gilts, total cost reduction by two phase feeding compared to single feeding was 9.1%, but in barrows it was just 3.19%. Relative margin increased with two phase feeding by 2.5% in gUts and 0.2% in barrows. There was a tendency that backfat thickness at 10th rib of gilts was thinner than that of barrows (p>0.05). Within the same sexes, there was no treatment effect on back fat thickness (p>0.05). Carcass grade was improved by two phase feeding compared to single feeding. Carcass grade of gilts was significantly better than that of barrows (p<0.001). From this results, it is concluded that finishing pigs could be fed two-phase diets to improve profit and reduce pollution.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1997년도 춘계학술대회
• /
• pp.358-361
• /
• 1997

This paper describes a multichannel epileptic seizure detection algorithm based on wavelet transform(WT), artificial neural network(ANN) and expert system. First, through the WT, a small number of wavelet coefficients is used to represent the single channel epileptic spike. Next, 3-layer feed-forward network employing the error back propagation algorithm is trained and tested using parameters obtained above. Finally, 16 channel expert system which is based on clinical experience is introduced as a artifact rejection and reliable detection. The suggested algorithm was implemented on personal computer(PC). Two main events i.e., epileptiform and normal activities, were selected from 32 person's EEGs(normal: 20, seizure disorder: 12) in consensus among experts. The result was that WT reduced data input size and ANN detected 97 of the 100 EEGs containing definite spike - sensitivity of 97%. Expert rule system was capable of rejecting a wide variety of artifacts commonly found in EEG recordings. It also reduced false positive detections of ANN.