• Steel and Composite Structures
• /
• v.49 no.6
• /
• pp.633-643
• /
• 2023

In this study, the natural frequencies of Functional Graded Materials (FGM) plates are predicted using Artificial Neural Network (ANN). A model based on Third-order Shear Deformation Theory (TSDT) and FEM is used to train the ANN model. Different training methods are tested to simulate input and output dependency. As this is a parametric model, several architectures and optimization algorithms were tested. The proposed model allows us to minimize the CPU time to evaluate candidate material properties for FGM plate material selection and demonstrate their influence on dynamic behavior. Consequently, the time required for the FGM design process (candidate materials for material selection) and the geometric optimization of the FGM structure would remain reasonable. The ANN model can help industries to produce FGM plates with good mechanical properties of the selected materials. I addition, this model can be used to directly predict vibration behavior by testing a large number of FGM plates, representing all possible combinations of metals and ceramics in today's industry, without having to solve any eigenvalue problems.

• Proceedings of the KIEE Conference
• /
• 2008.10b
• /
• pp.239-240
• /
• 2008

본 논문에서는 Subtractive clustering 알고리즘을 이용한 Fuzzy Radial Basis Function Neural Network (FRBFNN)의 규칙 수를 자동적으로 생성하는 방법을 제시한다. FRBFNN은 멤버쉽 함수로써 기존 RBFNN에서 가우시안이나 타원형 형태의 특정 RBF를 사용하는 구조와 달리 Fuzzy C-Means clustering 알고리즘에서 사용하는 거리에 기한 멤버쉽 함수를 사용하여 전반부의 공간 분할 및 활성화 레벨을 결정하는 구조이다. 본 논문에서는 데이터의 밀집도에 기반을 두어 클러스터링을 하는 Subtractive clustering 알고리즘을 사용하여 퍼지 규칙의 수와 같은 의미를 갖는 분할할 입력공간의 수와 분할된 입력공간의 중심값을 동정하며, Least Square Estimator (LSE) 알고리즘을 사용하여 후반부 다항식의 계수를 추정 한다.

• 제어로봇시스템학회:학술대회논문집
• /
• 1994.10a
• /
• 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.

• Proceedings of the KIEE Conference
• /
• 1998.07b
• /
• pp.566-568
• /
• 1998

In this paper, we suggest a optimal design method of Fuzzy-Neural Networks model for complex and nonlinear systems. FNNs have the stucture of fusion of both fuzzy inference with linguistic variables and Neural Networks. The network structure uses the simpified inference as fuzzy inference system and the BP algorithm as learning procedure. And we use a clustering algorithm to find initial parameters of membership function. The parameters such as membership functions, learning rates and momentum coefficients are easily adjusted using the genetic algorithms. Also, the performance index with weighted value is introduced to achieve a meaningful balance between approximation and generalization abilities of the model. To evaluate the performance index, we use the time series data for gas furnace and the sewage treatment process.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.11
• /
• pp.2103-2108
• /
• 2011

In this paper, the problem of passivity analysis for neural networks with time-varying delays and norm-bounded parameter uncertainties is considered. By constructing a new augmented Lyapunov functional, a new delay-dependent passivity criterion for the network is established in terms of LMIs (linear matrix inequalities) which can be easily solved by various convex optimization algorithms. Two numerical example are included to show the effectiveness of proposed criterion.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.9
• /
• pp.1809-1814
• /
• 2009

In this paper, the problem of stability analysis for neural networks with discrete and distributed time-varying delays is considered. By constructing a new Lyapunov functional, a new delay-dependent stability criterion for the network is established in terms of LMIs (linear matrix inequalities) which can be easily solved by various convex optimization algorithms. Two numerical example are included to show the effectiveness of proposed criterion.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.88.5-88
• /
• 2001

We propose a neural network solver for an inverse problem. The problem is that input data with complete teaching include defects and predict the defect value. The solver is constructed of a three layer neural network whose learning method is combined from BP and reconstruction learning. The input data for the defects are unknown; therefore, the circulation of an arithmetic progression replaces them; rightly, the learning procedure is not converged for the circulation data vut for the normal data. The learning is quitted after such a learning status id kept. Then, we search a minimum of the differences between teaching data and output of the circulation. Then, we search a minimum of the ...

• Geomechanics and Engineering
• /
• v.2 no.1
• /
• pp.57-69
• /
• 2010

A major consideration in the design of tunnels in urban areas is the prediction of the ground movements and surface settlements associated with the tunneling operations. Excessive ground movements can damage adjacent building and utilities. In this paper, a neural network model is used to predict the maximum surface settlement, based on instrumented results from three separate EPB tunneling projects in Singapore. This paper demonstrates that by coupling the trained neural network model to a spreadsheet optimization technique, the reliability assessment of the settlement serviceability limit state can be carried out using the first-order reliability method. With this method, it is possible to carry out sensitivity studies to examine the effect of the level of uncertainty of each parameter uncertainty on the probability that the serviceability limit state has been exceeded.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.27 no.2
• /
• pp.377-386
• /
• 2021

The A60 class deck penetration piece is a fire-resistant system installed on a horizontal compartment to prevent flame spreading and protect lives in fire accidents in ships and offshore plants. This study deals with approximate optimization using discrete variables for the fire resistance design of an A60 class deck penetration piece using different surrogate models and a genetic algorithm. Transient heat transfer analysis was performed to evaluate the fire resistance design of the A60 class deck penetration piece. For the approximate optimization of the piece, the length, diameter, material type, and insulation density were applied to discrete design variables, and temperature, productivity, and cost constraints were considered. The approximate optimum design problem based on the surrogate models was formulated such that the discrete design variables were determined by minimizing the weight of the piece subjected to the constraints. The surrogate models used in the approximate optimization were the response surface model, Kriging model, and radial basis function-based neural network. The approximate optimization results were compared with the actual analysis results in terms of approximate accuracy. The radial basis function-based neural network showed the most accurate optimum design results for the fire resistance design of the A60 class deck penetration piece.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.459-462
• /
• 2000

In this paper, we propose the initial optimized structure of the Radial Basis Function Network which is more simple in the part of the structure and converges more faster than Neural Network with the analysis method using Time-Frequency Localization. When we construct the hidden node with the Radial Basis Function whose localization is similar with an approximation target function in the plane of the Time and Frequency, we make a good decision of the initial structure having an ability of approximation.