• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.236-242
• /
• 1992

Multi-layered neural networks are applied to the identification and control of nonlinear dynamical system. Traditional adaptive control techniques can only deal with linear systems or some special nonlinear systems. A scheme for combining multi-layered neural networks with model reference network techniques has the capability to learn the nonlinearity and shows the great potential for adaptive control. In many interesting cases the system can be described by a nonlinear model in which the control input appears linearly. In this paper the identification of linear and nonlinear part are performed simultaneously. The projection algorithm and the new estimation method which uses the delta rule of neural network are compared throughout the simulation. The simulation results show that the identification and adaptive control schemes suggested are practically feasible and effective.

• Proceedings of the Korea Institutes of Information Security and Cryptology Conference
• /
• 2001.11a
• /
• pp.43-47
• /
• 2001

본 논문은 신경망 이론중 하나인 Adaptive Resonance Theory(ART)을 사용하여 네트워크 상의 불법적인 침입을 탐지하는 기법에 대한 연구이다. ART는 비교사 학습을 하는 신경망으로써, 적응적인 학습능력이 있으며, 또 새로운 패턴에 대해서 새로운 클러스터를 생산하는 능력이 있다. ART의 이러한 특성을 이용하여, 여러 가지 침입패턴을 네트워크상에서 생산하여 학습을 시키고, 또 test 했으며, test 이후에도 on-line 상에서 새로운 공격 pattern도 찾아냄을 보였다. 따라서, 이미 알려진 침입뿐만 아니라 새롭게 발생하는 침입 기법에 대해서도 새로운 rule의 첨가 없이 적극적으로 대처할 수 있을 것으로 예측된다.

• Proceedings of the KIEE Conference
• /
• 1998.07c
• /
• pp.896-898
• /
• 1998

Recent development of the digital computer and communication technology has made the concept of the adaptive protection possible, which is to adapt the operating parameters of the protective devices to the system changes, so that the best protection function can be maintained all the time. In order to achieve the adaptive protection, it is necessary to have the way to determine whether the change of the settings is needed under the certain system change or how good the current protection level is. This paper proposed the protectability index, which is a way to evaluate the protection level of the system under arbitary conditions and the operating strategy of the adaptive protection utilizing this index. It is based on an hierachical evaluation model and the evidence combination rule of the Dempster-Shafer theory.

• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2292-2295
• /
• 2001

In this paper, The control problem for a series DC motor is considered to adaptive fuzzy sliding-mode control scheme. Based on a nonlinear mathematical model of a series connected DC motor, instead of the combination of a nonlinear transformation and state feedback(feedback linearization) reduces the nonlinear control design. To demonstrate its effectiveness, an experimental study of this controller is presented. Two sets of fuzzy rule bases are utilized to represent the equivalent control input with unknown system functions of the main target. The membership functions of the THEN-part, which is used to construct a suitable equivalent control of SMC, are changed according to the adaptive law. With such a design scheme, we not only maintain the distribution of membership functions over state space but also reduce computing time considerably.

• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2081-2083
• /
• 2001

In this paper, without mathematical modeling dynamics, the plant parameter in sliding mode are estimated by the indirect adaptive fuzzy control. Adaptive laws for fuzzy parameters and fuzzy rule structure are established so that the whole system is stable in the sense of Lyapunov stability. The computer simulation results for inverted pendulum system show the performance of the proposed fuzzy sliding mode controller.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.6 no.2
• /
• pp.43-51
• /
• 1996

Fuzzy rules in fuzzy logic control play a major role in deciding the control dynamics of a fuzzy logic controller. Thus, control performance is mainly determined by the quality of fuzzy rules. This paper introduces an optimization method for fuzzy rules using GAS with adaptive probabilies of crossover and mutation. Also we design two fitness measures to satisfy control objectives by partitioning the response of a plant into two parts. An initial population is generated by an automatic fuzzy rule generation method instead of random selection for fast a.pproaching to the final solution. We employed a nonlinear plant to simulate our method. It is shown through simulation that our method is reasonable and can be useful for optimizing fuzzy rules.

• KIEAE Journal
• /
• v.14 no.3
• /
• pp.97-103
• /
• 2014

This study aimed at developing integrated logic for controlling heating device and openings of the double skin facade buildings. Two major logics were developed-rule-based control logic and artificial neural network based control logic. The rule based logic represented the widely applied conventional method while the artificial neural network based logic meant the optimal method. Applying the optimal method, the predictive and adaptive controls were feasible for supplying the advanced thermal indoor environment. Comparative performance tests were conducted using the numerical computer simulation tools such as MATLAB (Matrix Laboratory) and TRNSYS (Transient Systems Simulation). Analysis on the test results in the test module revealed that the artificial neural network-based control logics provided more comfortable and stable temperature conditions based on the optimal control of the heating device and opening conditions of the double skin facades. However, the amount of heat supply to the indoor space by the optimal method was increased for the better thermal conditioning. The number of on/off moments of the heating device, on the other hand, was significantly reduced. Therefore, the optimal logic is expected to beneficial to create more comfortable thermal environment and to potentially prevent system degradation.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.5
• /
• pp.672-676
• /
• 2010

This paper presents a learning rule that weights more on data near decision boundary. This learning rule generates better decision boundary by reducing the effect of outliers on the decision boundary. The proposed learning rule is integrated into IAFC neural network. IAFC neural network is stable to maintain previous learning results and is plastic to learn new data. The performance of the proposed fuzzy neural network is compared with performances of LVQ neural network and backpropagation neural network. The results show that the performance of the proposed fuzzy neural network is better than those of LVQ neural network and backpropagation neural network.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.10a
• /
• pp.361-367
• /
• 1998

The main two applications of the Genetic Algorithms(GA) are the optimization and the machine learning. Machine Learning has two objectives that make the complex system learn its environment and produce the proper output of a system. The machine learning using the Genetic Algorithms is called GA machine learning or genetic-based machine learning (GBML). The machine learning is different from the optimization problems in finding the rule set. In optimization problems, the population of GA should converge into the best individual because optimization problems, the population of GA should converge into the best individual because their objective is the production of the individual near the optimal solution. On the contrary, the machine learning systems need to find the set of cooperative rules. There are two methods in GBML, Michigan method and Pittsburgh method. The former is that each rule is expressed with a string, the latter is that the set of rules is coded into a string. Th classifier system of Holland is the representative model of the Michigan method. The classifier systems arrange the strength of classifiers of classifier list using the message list. In this method, the real time process and on-line learning is possible because a set of rule is adjusted on-line. A classifier system has three major components: Performance system, apportionment of credit system, rule discovery system. In this paper, we solve the food search problem with the learning and evolution of an artificial ant using the learning classifier system.

• Journal of Intelligence and Information Systems
• /
• v.15 no.4
• /
• pp.123-140
• /
• 2009

Fuzzy neural network is an integrated model of artificial neural network and fuzzy system and it has been successfully applied in control and forecasting area. Recently ANFIS(Adaptive Network-based Fuzzy Inference System) has been noticed widely among various fuzzy neural network models because of outstanding performance of control and forecasting accuracy. ANFIS has capability to refine its fuzzy rules interactively with human expert. In particular, when we use initial rule structure for machine learning which is generated from human expert, it is highly probable to reach global optimum solution as well as shorten time to convergence. We propose metrics to evaluate interpretability of generated rules as a means of acquiring domain knowledge and compare level of interpretability of ANFIS fuzzy rules to those of C5.0 classification rules. The proposed metrics also can be used to evaluate capability of rule generation for the various machine learning methods.