• Proceedings of the Korean Nuclear Society Conference
• /
• 1996.11a
• /
• pp.73-78
• /
• 1996

In this study, we proposed a fuzzy gain scheduler with intelligent learning algorithm for a reactor control. In the proposed algorithm, we used the gradient descent method to learn the rule bases of a fuzzy algorithm. These rule bases are learned toward minimizing an objective function, which is called a performance cost function. The objective of fuzzy gain scheduler with intelligent learning algorithm is the generation of adequate gains, which minimize the error of system. The condition of every plant is generally changed as time gose. That is, the initial gains obtained through the analysis of system are no longer suitable for the changed plant. And we need to set new gains, which minimize the error stemmed from changing the condition of a plant. In this paper, we applied this strategy for reactor control of nuclear power plant (NPP), and the results were compared with those of a simple PI controller, which has fixed gains. As a result, it was shown that the proposed algorithm was superior to the simple PI controller.

• Proceedings of the KIEE Conference
• /
• 1995.11a
• /
• pp.207-209
• /
• 1995

The PID controller is widely used due to its fast response and robustness. But its performance is not so good compared with modem controllers such as adaptive, robust, fuzzy, neural controller. Therefore, it is natural to replace PID controller by modem controllers. But, the problem is that modem controller can not be easily applied to the real time process. Hence, this paper proposes such a structure that PID controller and Self-Learning Fuzzy Controller(SLFC) are in parallel with each other. The parameter of SLFC will be updated by gradient descent method using neuro - identifier. The usefulness of this hybrid controller will be proved by simulation results.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.70-70
• /
• 2000

Recently, a neuro-fuzzy approach, a combination of neural networks and fuzzy reasoning, has been playing an important role in the motor control. In this paper, a novel method of fiction compensation using neuro-fuzzy architecture has been shown to significantly improve the performance of a DC motor system with nonlinear friction characteristics. The structure of the controller is the neuro-fuzzy network with the TS(Takagi-Sugeno) model. A back-propagation neural network based on a gradient descent algorithm is employed, and all of its parameters can be on-line trained. The performance of the proposed controller is compared with both a conventional neuro-controller and a PI controller.

• Proceedings of the Korean Society of Soil and Groundwater Environment Conference
• /
• 2006.04a
• /
• pp.391-393
• /
• 2006

In this study, the performance of artificial neural network (ANN) models for estimating spatial distribution of water quality was evaluated using electric conductivity (EC) values in landfill site. For the ANN model development, feedforward neural networks and backpropagation algorithm with gradient descent method were used. In Test 1, the interpolation ability of the ANN model was evaluated. Results of the ANN model were more precise than those of the Kriging model. In Test 2, spatial distributions of EC values were predicted using precipitation data. Results seemed to be reasonable, however, they showed a limitation of ANN models in extrapolations.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.796-798
• /
• 2005

본 연구에서는 엔트로피 이론을 사용하여 ICA(Independent Component Analysis) 점수함수를 생성하는 새로운 밀도추정자(Density Estimator)를 제안한다. 원 신호에 대한 밀도함수의 추정은 적당한 점수함수를 생성하기 위해 필요하고, 미분 가능한 밀도함수인 커널을 이용한 밀도추정법(Kernel Density Estimation)을 이용하여 점수함수를 생성하였다. 보다 빠른 점수함수의 생성을 위해서 식의 형태를 convolution 형태로 표현하였으며, ICA 학습을 위해서 결합엔트로피를 최대화(Joint Entropy Maximization)하는 방향으로 커널의 폭을 학습하였다. 이를 위해서 기울기 강하법(Gradient descent method)를 사용하였으며, 이러한 제약 사항은 새로운 밀도 추정자를 설계하기 위한 기본적인 개념을 나타낸다. 실험결과, 커널의 폭을 담당하는 smoothing parameters들이 일정한 값으로 학습함을 알 수 있었다.

• Proceedings of the KIEE Conference
• /
• 1997.07b
• /
• pp.394-396
• /
• 1997

This paper describes RBF network controller based on fuzzy control rules for intelligent control of nonlinear systems. The proposed scheme is derived from the functional equivalence between RBF networks and fuzzy inference systems. The design procedure of the proposed scheme is realized by first transforming the fuzzy control rules into the parameters of RBF networks. The optimized RBF network controller is then performed through the gradient descent learning mechanism to an error function. The proposed method is rigorously tested using a nonlinear and unstable nonlinear system. Simulation is performed to demonstrate the feasibility and effectiveness of the proposed scheme.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.8
• /
• pp.211-218
• /
• 1994

In this paper we develope a hybrid learning rule to improve the robustness of multi-layer Perceptions. In most neural networks the activation of a neuron is deternined by a nonlinear transformation of the weighted sum of inputs to the neurons. Investigating the behaviour of activations of hidden layer neurons a new learning algorithm is developed for improved robustness for multi-layer Perceptrons. Unlike other methods which reduce the network complexity by putting restrictions on synaptic weights our method based on error-backpropagation increases the complexity of the underlying proplem by imposing it saturation requirement on hidden layer neurons. We also found that the additional gradient-descent term for the requirement corresponds to the Hebbian rule and our algorithm incorporates the Hebbian learning rule into the error back-propagation rule. Computer simulation demonstrates fast learning convergence as well as improved robustness for classification and hetero-association of patterns.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.10a
• /
• pp.560-567
• /
• 1998

This paper concerned about a imputed color printed material image in camera to decrease noise and distortion by processing median filtering with input image to identical condition. Also this paper proposed the way of compares a normal printed material with an abnormal printed material color tone with trained a learning of the error back-propagation to block classification by extracting five place from identical block(3${\times}$3) of color printed material R, G, B value. As a representative algorithm of multi-layer perceptron the error Back-propagation technique used to solve complex problems. However, the Error Back-propagation is algorithm which basically used a gradient descent method which can be converged to local minimum and the Back Propagation train include problems, and that may converge in a local minimum rather than get a global minimum. The network structure appropriate for a given problem. In this paper, a good result is obtained by improve initial condition and adjust th number of hidden layer to solve the problem of real time process, learning and train.

• Journal of the Korean Mathematical Society
• /
• v.61 no.2
• /
• pp.395-408
• /
• 2024

There have been numerous studies on the characteristics of the solutions of ordinary differential equations for optimization methods, including gradient descent methods and alternating direction methods of multipliers. To investigate computer simulation of ODE solutions, we need to trace pseudo-orbits by real orbits and it is called shadowing property in dynamics. In this paper, we demonstrate that the flow induced by the alternating direction methods of multipliers (ADMM) for a C² strongly convex objective function has the eventual shadowing property. For the converse, we partially answer that convexity with the eventual shadowing property guarantees a unique minimizer. In contrast, we show that the flow generated by a second-order ODE, which is related to the accelerated version of ADMM, does not have the eventual shadowing property.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.11 no.1
• /
• pp.82-96
• /
• 2019

A semi-submersible structure has been widely used for offshore drilling and production of oil and gas. The small water plane area makes the structure very sensitive to weight increase in terms of payload and stability. Therefore, it is necessary to lighten the substructure from the early design stage. This study aims at an optimization of hull structure based on a sophisticated yield and buckling strength in accordance with classification rules. An in-house strength assessment system is developed to automate the procedure such as a generation of buckling panels, a collection of required panel information, automatic buckling and yield check and so on. The developed system enables an automatic yield and buckling strength check of all panels composing the hull structure at each iteration of the optimization. Design variables are plate thickness and stiffener section profiles. In order to overcome the difficulty of large number of design variables and the computational burden of FE analysis, various methods are proposed. The steepest descent method is selected as the optimization algorithm for an efficient search. For a reduction of the number of design variables and a direct application to practical design, the stiffener section variable is determined by selecting one from a pre-defined standard library. Plate thickness is also discretized at 0.5t interval. The number of FE analysis is reduced by using equations to analytically estimating the stress changes in gradient calculation and line search steps. As an endeavor to robust optimization, the number of design variables to be simultaneously optimized is divided by grouping the scantling variables by the plane. A sequential optimization is performed group by group. As a verification example, a central column of a semi-submersible structure is optimized and compared with a conventional optimization of all design variables at once.