• 한국물환경학회지
• /
• 제21권1호
• /
• pp.21-28
• /
• 2005

The training and prediction modeling using an artificial neural network was implemented to predict the turbidity of treated water as well as to estimate the optimized feed concentration of polyaluminium chloride (PACl) in a water treatment plant. The parameters used in the input layers were pH, temperature, turbidity and alkalinity, while those in output layers were PACl and turbidity of treated water. Levenberg-Marquadt method of feedforward back-propagation perceptron in the neural network toolbox of MATLAB program was used in this study. Correlation coefficients of the training data with the measured data were 0.9997 for PACl and 0.6850 for turbidity and those of the testing data with measured data were 0.9140 for PACl and 0.3828 for turbidity, when four parameters at input layer, 12-12 nodes each at both the first and the second hidden layers, and two parameters(PACl and turbidity) at output layer were used. Although the predictability of PACl was improved, compared to that of the previous studies to use the only coagulant dose as output layer, turbidity in treated water could not be predicted well. Acquisition of more data through several years obtained with the advanced on-line measuring system could make the artificial neural network useful and practical in actual water treatment plants.

• 한국컴퓨터산업학회논문지
• /
• 제6권3호
• /
• pp.421-428
• /
• 2005

최근의 축산업은 대규모화와 자동화가 급진전 되면서 해당농장의 여건에 맞는 맞춤사료 개발시스템 구축이 절실하다. 오프라인상의 해당 각 농장들은 이에 대한 시스템 구축비용 등으로 인하여 많은 어려움을 겪고 있는 실정이다. 본 논문에서는 IT기술의 디지털 정보와 기술을 축산현장에 접목하는 시스템으로써 인터넷 서비스 환경에서 해당 양돈 농장에 맞춤화한 사료를 공급할 수 있는 시스템을 개발한다. 이 시스템은 경제적인 사료공급 뿐만 아니라 사료 생산비의 효율적인 운영 등 농${\cdot}$축산업의 업무 환경에 적합한 맞춤사료 DB구축을 가능케 하며 사육비 절감 등의 이점이 있다. 또한, 농장의 생산에 관련된 장치와 요소들을 디지털화하고 네트워크 환경을 구축하여 실시간대로 확인할 수 있는 21세기형 디지털 농${\cdot}$축산 솔루션으로써 기대 된다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 하계종합학술대회논문집
• /
• pp.917-920
• /
• 1998

In this paper, we propose a modified elman neural network structure for nonlinear system identification. The proposed structure is that all of network output feed back into hidden units and output units. Learning algorithm is standard back-propagation algorithm. The simulation showed the effectiveness of using the modified elman neural network structure in the nonlinear system identification.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2003년도 춘계학술대회 논문집
• /
• pp.272-277
• /
• 2003

In order to predict vibration during end-milling process, the cutting dynamics was modelled by using neural network and combined with structural dynamics by considering dynamic cutting states. Specific cutting constants of the cutting dynamics model were obtained by averaging cutting forces and tool diameter, cutting speed, feed, axial depth radial depth were considered as machining factors. Cutting farces by test and by neural network simulation were compared and the vibration during end-milling was simulated.

• Smart Structures and Systems
• /
• 제18권6호
• /
• pp.1233-1250
• /
• 2016

Cracks in plate-like structures are some of the main reasons for destruction of the entire structure. In this study, a novel two-stage methodology is proposed for damage detection of flexural plates using an optimized artificial neural network. In the first stage, location of damages in plates is investigated using curvature-moment and curvature-moment derivative concepts. After detecting the damaged areas, the equations for damage severity detection are solved via Bat Algorithm (BA). In the second stage, in order to efficiently reduce the computational cost of model updating during the optimization process of damage severity detection, multiple damage location assurance criterion index based on the frequency change vector of structures are evaluated using properly trained cascade feed-forward neural network (CFNN) as a surrogate model. In order to achieve the most generalized neural network as a surrogate model, its structure is optimized using binary version of BA. To validate this proposed solution method, two examples are presented. The results indicate that after determining the damage location based on curvature-moment derivative concept, the proposed solution method for damage severity detection leads to significant reduction of computational time compared with direct finite element method. Furthermore, integrating BA with the efficient approximation mechanism of finite element model, maintains the acceptable accuracy of damage severity detection.

• Journal of Electrical Engineering and Technology
• /
• 제10권2호
• /
• pp.634-640
• /
• 2015

This paper deals with the design of a microstrip line feed toppled T shaped microstrip patch antenna that gives dualband characteristics at 4 GHz and 6.73 GHz respectively. The simulation of proposed antenna geometry has been performed using method of moment based IE3D simulation software. A radial basis function neural network (RBFNN) is used for the estimation of bandwidth for dualband at 4 GHz and 6.73 GHz respectively. In RBFNN model, antenna parameters such as dielectric constant, height of substrate, and width are used as input and bandwidth of first and second band is considered as output of the network. To validate the RBFNN output, an antenna has been physically fabricated on glass epoxy substrate. The fabricated antenna can be utilized in S and C bands applications. RBFNN results are found in close agreement with simulated and experimental results.

• 전기학회논문지
• /
• 제60권9호
• /
• pp.1700-1707
• /
• 2011

Interior permanent magnet synchronous motor(IPMSM) adjustable speed drives offer significant advantages over induction motor drives in a wide variety of industrial applications such as high power density, high efficiency, improved dynamic performance and reliability. Since the fuzzy neural network(FNN) is recognized general approximate method to control non-linearities and uncertainties, the development of FNN control systems have also grown rapidly. The FNN controller is compounded of fuzzy and neural network. It has an advantage that is the robustness of fuzzy control and the ability to adapt of neural network. However, the FNN has static problem due to their feed-forward network structure. This paper proposes high performance speed control of IPMSM drive using the recurrent FNN(RFNN) which improved conventional FNN controller. The RFNN has excellent dynamic response characteristics because of it has internally feed-back structure. Also, this paper proposes speed estimation of IPMSM drive using ANN. The proposed method is analyzed and compared to conventional FNN controller in various operating condition such as parameter variation, steady and transient states etc.

• 한국산학기술학회논문지
• /
• 제20권12호
• /
• pp.663-670
• /
• 2019

터빈 사이클 보정 열 성능 분석은 발전소의 현재 성능을 결정하고 향상된 경제성 운전을 위해 요구된다. 본 연구에서는 신뢰성있는 성능 분석을 위해서 산업 표준인 ASME(American Society of Mechanical Engineers) PTC(Performance Test Code)를 기본으로 성능 분석에서 우선적으로 중요하게 적용되는 주급수 유량을 대상으로 영역별 판정 알고리즘을 개발하고 각 영역별로 현재의 터빈 사이클 성능을 추정하는 알고리즘을 개발하였다. 추정 알고리즘은 측정 상태량의 상관 관계를 기반으로 영역별로 형상 분류를 제시하고, 이를 기반으로 커널 회귀 모델을 이용하여 학습된 추정 모델을 구성하였으며, 커널 회귀 모델링의 우수성을 검증하기 위하여 신경 회로망 모델의 학습 결과와 비교하였다. 주급수 유량의 형상 특성에 따른 분류 및 추정 모델은 터빈 사이클에서 정확한 보정 열 성능 분석을 제공함으로써 성능 분석의 신뢰도를 증가시킬 수 있었으며 다른 성능 결정 변수에 대한 학습 및 검증 모델로 사용될 수 있다.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2004년도 춘계학술대회 논문집
• /
• pp.381-386
• /
• 2004

A flexible disk grinding system process has been introduced that utilized varying disk orientation with respect to workpiece along with the applied feed speed. A known process model methodologies has been used to fomulate processed surface profiles. Various process conditions including cutting speed, maximum feed speed and orientation angles could applied to observe process outcomes. Even though continuous and constant feed speed has been applied to the process, the results from the trapezoidal input velocity profiles would be observed and compared. Based on the control strategies including neural network methodologies, several output results were compared to find the optimum process condition.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2004년도 추계학술대회 논문집
• /
• pp.309-314
• /
• 2004

A flexible disk grinding system process has been introduced that utilized varying disk orientation with respect to workpiece along with the applied feed speed. Various process conditions including cutting speed, maximum feed speed and orientation angles could applied to observe process outcomes. Even though continuous and constant feed speed has been applied to the process, the results from the trapezoidal input velocity profiles would be observed and compared. Based on the control strategies including neural network methodologies, several output results were compared to find the optimum process condition. Two axis control results were displayed showing better performance with higher trajectory error for larger training epoch.