• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.43 no.6 s.348
• /
• pp.53-61
• /
• 2006

In high speed flash type or pipelining type A/D Converter, the faster sampling frequency is, the more the effect of DC reference fluctuation is increased by clock feed-through and kick-back. When we measure A/D Converter, further, external noise increases reference voltage fluctuation. Thus reference fluctuation reduction circuit must be needed in high speed A/D converter. Conventional circuit simply uses capacitor but layout area is large and it's not efficient. In this paper, a reference fluctuation reduction circuit using transmission gate is proposed. In order to verify the proposed technique, we designed and manufactured 6bit 2GSPS CMOS A/D converter. The A/D converter is based on 0.18um 1-poly 5-metal N-well CMOS technology, and it consumes 145mW at 1.8V power supply. It occupies chip area of $977um\times1040um$. Experimental result shows that SNDR is 36.25 dB and INL/DNL ${\pm}0.5LSB$ when sampling frequency is 2GHz.

• Asian-Australasian Journal of Animal Sciences
• /
• v.18 no.10
• /
• pp.1479-1484
• /
• 2005

A study was conducted to determine the effects of restricted feeding and feed form on the growth performance, characteristics of carcass and digestive tract, and days to first egg of Japanese quail (JQ, Coturnix coturnix japonica). A total of 240 oneweek-old JQ chicks were allocated randomly into 4 experimental groups that consisted of 3 replicates according to a 2${\times}$2 factorial arrangement for two feeding methods (ad libitum, AF and restricted feeding, RF) and two diet forms (mash, MD and crumble, CD). The JQ chicks were placed in a room with floor battery brooders and fed a commercial starter diet from 7 to 14 d of age. According to the experimental design, four treatments (1: ad libitum MD, 2: restricted MD, 3: ad libitum CD, and 4: restricted CD) were applied. Feed restriction was applied by 30% reduction of ad libitum feed intake for both MD and CD from 15 to 28 d of age. All birds were fed ad libitum with treatment diets from 29 d of age until the first laid egg seen (45 d of age). The commercial starter diet, MD and CD were in the same nutrient content (240 g crude protein with 13.4 MJ ME per kg diet). The body weight and overall feed conversion ratio (g feed/g gain) were higher (p<0.05) for the AF quails than the RF at 42 d of age. Carcass weights, dressing percentage and percentage yields of breast and back were similar for AF and RF groups at 42 d of age. The RF delayed (p<0.05) onset of egg production 2 days compared to the AF. Quail fed with the CD showed higher value (p<0.05) for carcass weight and dressing percentage at 42 d of age compared to birds fed with the MD. The interaction effect of feeding method${\times}$feed form on any of the studied parameters was not significant. The results suggest that feed restriction as in the present study can achieve a better feed conversion without reduction in carcass weight, and a significant benefit of feeding the crumble diet over the mash diet was obtained in terms of carcass weight in the JQ.

• Korean Journal of Agricultural Science
• /
• v.42 no.3
• /
• pp.253-259
• /
• 2015

This study was conducted to investigate the effect of different additives on the growth performance, feed efficiency and carcass characteristics in Holstein steers during 18month fattening periods. Twenty four Holstein steers, 5months of age and 176.6kg, were randomly allocated to 3 experimental groups 8 animals each for 18-months feeding trial. The groups were control (not additive), T1 (fed zinc, Vitamin C) and T2 (fed zinc, Vitamin C, Vitamin B6 and Selenium). According to feeding additives, final weight was not significantly different among the treatment groups, tended to be high at T2 group (827kg) compared to the other groups. Average daily gain was not different among the treatment groups during the experimental periods, but T2 group was significantly greater than T1 group in growing stage (p<0.05). The feed additives had no effects on DMI during experimental periods. Feed conversion ratio of T1 group in growing stage was significantly higher than those of other groups (p<0.05), average feed conversion ratio was tend to be decreased at T2 group rather compared with other groups. In the results of yield traits, carcass weight were relatively higher in T2 group than other groups (p<0.05). Rib-eye area, back fat thickness and yield index were similar between groups. In quality traits, marbling, meat color, fat color, texture and maturity were not significantly different among the groups. In economic efficiency, income was highest at T2 group as 91~393 thousand won among 3 groups.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.28 no.3
• /
• pp.254-262
• /
• 2008

The objective of the proposed study is to produce a tool-condition monitoring (TCM) strategy that will lead to a more efficient and economical drilling tool usage. Drill-wear monitoring is an important attribute in the automatic cutting processes as it can help preventing damages of the tools and workpieces and optimizing the tool usage. This study presents the architectures of a multi-layer feed-forward neural network with back-propagation training algorithm for the monitoring of drill wear. The input features to the neural networks were extracted from the AE signals using the wavelet transform analysis. Training and testing were performed under a moderate range of cutting conditions in the dry drilling of steel plates. The results indicated that the extracted input features from AE signals to the supervised neural networks were effective for drill wear monitoring and the output of the neural networks could be utilized for the tool life management planning.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1995.09a
• /
• pp.325-333
• /
• 1995

Back-propagation is the most widely used algorithm for supervised learning in multi-layer feed-forward networks. However, back-propagation is very slow in convergence. In this paper, a new weight initialization method, called rough map initialization, in multi-layer perceptrons is proposed. To overcome the long convergence time, possibly due to the random initialization of the weights of the existing multi-layer perceptrons, the rough map initialization method initialize weights by utilizing relationship of input-output features with singular value decomposition technique. The results of this initialization procedure are compared to random initialization procedure in encoder problems and xor problems.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.50.6-50
• /
• 2002

In this paper, the relationship between a slave robot and the uncertain remote environment is modeled as the impedance to generate the virtual force to feed back to the operator. For the control of a teleoperated mobile robot equipped with camera, the teleoperated mobile robot take pictures of remote environment and sends the visual information back to the operator over the Internet. Because of the limitation of communication bandwidth and narrow view-angles of camera, it is not possible to watch the environment clearly, especially shadow and curved areas. To overcome this problem, the virtual force is generated according to both the distance between the obstacle and robot and the approachin...

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.14 no.3
• /
• pp.65-73
• /
• 2015

This paper presents a method of predicting the machining parameters on the turning process of low carbon steel using a neural network with back propagation (BP) and particle swarm optimization (PSO). Cutting speed, feed rate, and depth of cut are used as input variables, while surface roughness and electric current consumption are used as output variables. The data from experiments are used to train the neural network that uses BP and PSO to update the weights in the neural network. After training, the neural network model is run using test data, and the results using BP and PSO are compared with each other.

• Proceedings of the KSR Conference
• /
• 2001.05a
• /
• pp.424-430
• /
• 2001

현대건설(주)는 사업 주관사로서 해외 업체와 대만고속전철 턴키 공사를 2000 년 1 월에 공동으로 수주하였다. 고속전철의 총 연장 길이는 약 326 km 이며, 정거장 10 개소, Depot 및 야적장으로 구성되어 있다. 이번에 수주한 공구는 2 개의 연속된 공구 (C230, C240) 이며, 본 논문은 총 연장 23.6km 인 C230 공구에 대한 설계 과정을 수록하였다. C230 공구는 NATM 터널 (6.2km), Cut-and-Cover 터널 (0.5km), 교량 (7.8km) 및 토공 구간 (9.1km)으로 구성되어 있다. 전 구간의 지반조건은 "매우" 취약한 매질로 구성되어 있으며, 층리나 절리는 거의 발달되어 있지 않다. 따라서 화약발파에 의한 터널 굴착은 기계식 굴착 (Back-hoe Excavation) 방법에 비하여 현실성이 없는 것으로 분석되었다. 취약한 지반에서 계측 결과를 기준으로 굴착 공간을 안전하게 유지할 수 있는 NATM 보강 설계가 현지 암반조건에 가장 이상적인 방법으로 제시되었다. 특히, NAT설계는 대형 아파트 지역과 파쇄대 및 지하수 침투 예상지역을 통과하기 위하여 계측에 의한 Feed-back 과정을 탄력적으로 적용하도록 계획하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.11a
• /
• pp.19-22
• /
• 2006

In this paper we present a vision based place recognition method which uses Bayesian method with feed back of image retrieval. Both Bayesian method and image retrieval method are based on interest features that are invariant to many image transformations. The interest features are detected using Harris-Laplacian detector and then descriptors are generated from the image patches centered at the features' position in the same manner of SIFT. The Bayesian method contains two stages: learning and recognition. The image retrieval result is fed back to the Bayesian recognition to achieve robust and confidence. The experimental results show the effectiveness of our method.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.11
• /
• pp.158-165
• /
• 1999

It has been established that a crack has an important effect on the dynamic behavior of a structure. This effect depends mainly on the location and depth of the crack. To identify the location and depth of a crack in a structure, a method is presented in this paper which uses hybrid neuro-genetic technique. Feed-forward multilayer neural networks trained by back-propagation are used to learn the input)the location and dept of a crack)-output(the structural eigenfrequencies) relation of the structural system. With this neural network and genetic algorithm, it is possible to formulate the inverse problem. Neural network training algorithm is the back propagation algorithm with the momentum method to attain stable convergence in the training process and with the adaptive learning rate method to speed up convergence. Finally, genetic algorithm is used to fine the minimum square error.