• 전자공학회논문지B
• /
• 제31B권7호
• /
• pp.170-178
• /
• 1994

The nearest neighbor classification rule is widely used because it is not only simple but the error rate is asymptotically less than twice Bayes theoretical minimum error. But the method basically use the whole training patterns as the reference vectors. so that both storage and classification time increase as the number of training patterns increases. LVQ(Learning Vector Quantization) resolved this problem by training the reference vectors instead of just storing the whole training patterns. But it is a heuristic algorithm which has no theoretic background there is no terminating condition and it requires a lot of iterations to get to meaningful result. This paper is to propose a new training method of the reference vectors. which minimize the given error function. The nearest neighbor network,the network version of the nearest neighbor classification rule is proposed. The network is funtionally identical to the nearest neighbor classification rule is proposed. The network is funtionally identical to the nearest neighbor classification rule and the reference vectors are represented by the weights between the nodes. The network is trained to minimize the error function with respect to the weights by the steepest descent method. The learning algorithm is derived and it is shown that the proposed method can adjust more reference vectors than LVQ in each iteration. Experiment showed that the proposed method requires less iterations and the error rate is smaller than that of LVQ2.

• 대한기계학회논문집A
• /
• 제26권3호
• /
• pp.436-444
• /
• 2002

The measurement of residual stresses by the hole-drilling method has been commonly used to evaluate residual stresses in structural members. In this method, eccentricity can usually occur between the hole center and rosette gage center. In this study, the error due to the hole eccentricity is predicted using the artificial neural network. The neural network has trained training examples of stress ratio, normalized eccentricity, off-centered direction and stress error using backpropagation learning process. The prediction results of the error using the trained neural network are good agreement with FE analyzed ones.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2001년도 춘계학술대회논문집A
• /
• pp.956-963
• /
• 2001

The measurement of residual stresses by the hole-drilling method has been commonly used to evaluate residual stresses in structural members. In this method, eccentricity can usually occur between the hole center and rosette gage center. In this study, the error due to the hole eccentricity is predicted using the artificial neural network. The neural network has trained training examples of stress ratio, normalized eccentricity, off-centered direction and stress error using backpropagation loaming process. The prediction results of the error using the trained neural network are good agreement with FE analyzed ones.

• 대한전기학회논문지
• /
• 제39권12호
• /
• pp.1306-1316
• /
• 1990

A controller using a multilayer neural network is proposed to the dynamic control of a PUMA 560 robot arm. This controller is developed based on an error back-propagation (BP) neural network. Since the neural network can model an arbitrary nonlinear mapping, it is used as a commanded feedforward torque generator. A Proportional Derivative (PD) feedback controller is used in parallel with the feedforward neural network to train the system. The neural network was trained by the current state of the manipulator as well as the PD feedback error torque. No a priori knowledge on system dynamics is needed and this information is rather implicitly stored in the interconnection weights of the neural network. In another experiment, the neural network was trained with the current, past and future positions only without any use of velocity sensors. Form this thim window of position values, BP network implicitly filters out the velocity and acceleration components for each joint. Computer simulation demonstrates such powerful characteristics of the neurocontroller as adaptation to changing environments, robustness to sensor noise, and continuous performance improvement with self-learning.

• 한국항행학회논문지
• /
• 제11권3호
• /
• pp.313-318
• /
• 2007

본 논문은 FDA(Fault Detection Agent)의 설계와 구축을 설명한다. FDA는 홈 네트워크 환경에서 멀티미디어 원격 교육을 위한 소프트웨어 오류를 감지하기에 적합한 에이전트이다. 이 시스템은 ED, ES로 구성되어 있다. ED는 홈 네트워크 환경에서 멀티미디어 원격 교육을 위하여 훅 킹 기법으로 오류를 감지하는 에이전트이다. ES는 홈 네트워크 환경에서 멀티미디어 원격 교육을 위하여 오류를 공유하는 에이전트이다. 멀티미디어 공동 작업 환경의 관점에서 오류 공유는 협동 작업에 참가하는 참가자에게 상호작용적으로 오류를 공유한다. 훅 킹 방법과 가로채기 방법과의 비교를 통하여 효율성 분석을 하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 통신소사이어티 추계학술대회논문집
• /
• pp.183-186
• /
• 2003

In this paper, performance evaluation of HNCP home network is using stochastic activity network models is proposed. HNCP is a home network protocol for controling and monitoring home appliances using power line communication. a CSMA/CA with packet drop method is used in HNCP MAC layer. Using the proposed stochastic activity network models. performances of HNCP home networks with error-free environment and error environment are evaluated.

• 한국측량학회지
• /
• 제26권1호
• /
• pp.17-26
• /
• 2008

우리나라 1등 수준망은 1974년부터 1986년간 관측된 자료를 1987년 동시 조정된 바 있다. 이후, 국토지리정보원에서는 2001년부터 2006년까지 전자레벨과 바코드 함척을 이용하여 1등 수준망에 대한 직접수준측량 실시하였고, 본 연구는 이들 관측자료 분석 및 수준망 조정 결과이다. 우리나라 1등 수준망은 총 36개 노선으로 구성되어 있는데, 그 중 34노선이 11개 폐합 환을 구성한다. 1등 수준측량 관측자료 중 4개 노선은 1등 수준측량 허용 왕복차를 초과하였으며, 3개 수준환은 허용 환 폐합차를 초과하였다. 수준점간 왕복차의 표준오차(${\eta}_1$)는 $0.2{\sim}1.7mm/{\surd}km$, 환 폐합차의 표준오차(${\eta}_2$)는 $2.0mm/{\surd}km$로 IGA의 고정밀 수준측량기준을 충족하였다. 1등 수준망은 1점(수준원점) 고정방식에 의한 망조정이 수행되었으며, 기준 표준오차($\hat{{\sigma}_0}$)는 $1.8mm/{\surd}km$로, 1987년 1등 수준망 조정의 기준 표준오차($\hat{{\sigma}_0}$)보다 2배 향상된 결과이다.

• 한국측량학회:학술대회논문집
• /
• 한국측량학회 2007년도 춘계학술발표회 논문집
• /
• pp.7-10
• /
• 2007

The 1st order vertical network of Korea was adjusted in 1987 at first time. This is the second adjustment of the 1st order vertical network of Korea by National Geographic Information Institute. All the levelling data were acquired by digital level with invar staff. The number of 1st order level lines are 36, and 34 level lines comprise 11 circles of level network. Backward and forward error of a few level lines are larger than the regulations of NGII, Korea. Also, 3 circles of vertical network has circuit closure error that is exceed the regulation. As the result of 1st order vertical network adjustment, the reference standard error of the vertical network was $1.8mm/{\surd}km$.

• 제어로봇시스템학회논문지
• /
• 제18권5호
• /
• pp.502-508
• /
• 2012

Heliostat, as a concentrator to reflect the incident solar energy to the receiver, is the most important system in the tower-type solar thermal power plant since it determines the efficiency and ultimately the overall performance of solar thermal power plant. Thus, a good sun tracking ability as well as a good optical property of it are required. Heliostat sun tracking system uses usually an open loop control system. Thus the sun tracking error caused by heliostat's geometrical error, optical error and computational error cannot be compensated. Recently use of sun tracking error model to compensate the sun tracking error has been proposed, where the error model is obtained from the measured ones. This work is a development of heliostat sun tracking error measurement and compensation method using BCS (Beam Characterization System). We first developed an image processing system to measure the sun tracking error optically. Then the measured error is modeled in linear polynomial form and neural network form trained by the extended Kalman filter respectively. Finally error models are used to compensate the sun tracking error. We also developed the necessary image processing algorithms so that the heliostat optical properties such as maximum heat flux intensity, heat flux distribution and total reflected heat energy could be analyzed. Experimentally obtained data shows that the heliostat sun tracking accuracy could be dramatically improved using either linear polynomial type error model or neural network type error model. Neural network type error model is somewhat better in improving the sun tracking performance. Nevertheless, since the difference between two error models in compensation of sun tracking error is small, a linear error model is preferred in actual implementation due to its simplicity.

• 제어로봇시스템학회논문지
• /
• 제16권7호
• /
• pp.711-719
• /
• 2010

Heliostat, as a concentrator reflecting the incident solar energy to the receiver located at the tower, is the most important system in the tower-type solar thermal power plant, since it determines the efficiency and performance of solar thermal plower plant. Thus, a good sun tracking ability as well as its good optical property are required. In this paper, we propose a method to compensate the heliostat sun tracking error. We first model the sun tracking error, which could be measured using BCS (Beam Characterization System), by multilayered neural network. Then the extended Kalman filter was employed to train the neural network. Finally the model is used to compensate the sun tracking errors. Simulated result shows that the method proposed in this paper improve the heliostat sun tracking performance dramatically. It also shows that the training of neural network by the extended Kalman filter provides faster convergence property, more accurate estimation and higher measurement noise rejection ability compared with the other training methods like gradient descent method.