• Journal of the Korean Society for Precision Engineering
• /
• v.14 no.4
• /
• pp.110-118
• /
• 1997

Multi-layer neural networks with error back-propagation algorithm have a great potential for identifying nonlinear systems with unknown characteristics. However, because they have a demerit that the speed of convergence is too slow, various methods for improving the training characteristics of backpropagition networks have been proposed. In this paper, a fuzzy division method is proposed to improve the convergence speed, which can find out an effective fuzzy division by the tuning of membership function and independently train each neural network after dividing the network model into several parts. In the simulations, the proposed method showed that the optimal fuzzy partitions could be found from the arbitray initial ones and that the convergence speed was faster than the traditional method without the fuzzy division.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.120.2-120
• /
• 2002

In this paper auto-tuning scheme of weight function in the neural networks has been suggested by immune algorithm for nonlinear process. A number of structures of the neural networks are considered as learning methods for control system. A general view is provided that they are the special cases of either the membership functions or the modification of network structure in the neural networks. On the other hand, since the immune network system possesses a self organizing and distributed memory, it is thus adaptive to its external environment and allows a PDP (parallel distributed processing) network to complete patterns against the environmental situation. Also. It can provi..

• Journal of the Korean Society for Precision Engineering
• /
• v.18 no.7
• /
• pp.46-52
• /
• 2001

This study describes the precision control of the torque standard machine using a self-tuning fuzzy controller. The torque standard machine should generate the accurate torque for calibrating a torque sensor. In order to reduce the relative expanded uncertainty of the torque standard machine, when a weight is hanged to the end of the torque arm for generating the torque, the sloped torque arm should be accurately controlled to the horizontal level. If the slope of the torque arm is larger from the inaccurate control, the uncertainty of the torque standard machine due to control will be larger. This applies the inaccurate torque to a torque sensor to calibrate, and the measuring error of the torque sensor generate from it. Therefore the torque arm of the torque standard machine is accurately controlled. In this paper, the self-tuning fuzzy controller was designed using a fuzzy theory, and the torque arm of the torque standard machine was accurately controlled. The control gain of the fuzzy controller, that is the membership function size of the error, the membership function size of the error change and the membership function size of the controller were determined from the self-tuning. The control results of the torque standard machine were the overshoot within 0.0076mm, the rise time within 16.70sec and the steady state error within 0.0076mm.

• The Proceedings of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.11 no.2
• /
• pp.102-109
• /
• 1997

The width of fuzzy membership function and control rule has an effect on performance of the fuzzy controller for electric equipment systems. In this paper, the neuro-fuzzy controller is proposed to im¬prove the performance of fuzzy controller. It has the width of membership function, that is adapted to the electrical parameter using multi-layer neural network, it is applied to first order electric power system with dead time and various plant constant. The related simulation resolts show that the pro¬posed neuro fuzzy controller are superior characteristics of improved performance

• Proceedings of the KIEE Conference
• /
• 2007.04a
• /
• pp.207-209
• /
• 2007

In this paper, we introduce a genetic optimization of fuzzy set-fuzzy model using successive tuning method to carry out the model identification of complex and nonlinear systems. To identity we use genetic alrogithrt1 (GA) sand C-Means clustering. GA is used for determination the number of input, the seleced input variables, the number of membership function, and the conclusion inference type. Information Granules (IG) with the aid of C-Means clustering algorithm help determine the initial paramters of fuzzy model such as the initial apexes of the, membership functions in the premise part and the initial values of polyminial functions in the consequence part of the fuzzy rules. The overall design arises as a hybrid structural and parametric optimization. Genetic algorithms and C-Means clustering are used to generate the structurally as well as parametrically optimized fuzzy model. To identify the structure and estimate parameters of the fuzzy model we introduce the successive tuning method with variant generation-based evolution by means of GA. Numerical example is included to evaluate the performance of the proposed model.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.12
• /
• pp.1101-1105
• /
• 2012

This study proposes a method for improvement of PD type fuzzy controller. The method includes self tuner using gradient algorithm that is one of the optimization algorithms. The proposed controller improves simple Takagi-Sugeno type FLC (Fuzzy Logic Control) system. The simple Takagi-Sugeno type FLC system changes nonlinear characteristic to linear parameters of consequent membership function. The simple FLC system could control the system by calibrating parameter of consequent membership function that changes the system response. While the determination on parameter of the simple FLC system works well only partially, the proposed method is needed to determine parameters that work for overall response. The simple FLC system doesn't predict the response characteristics. While the simple FLC system works just like proportional part of PID, our system includes derivative part to predict the next response. The proposed controller is constructed with P part and D part FLC system that characteristic parameter on system response is changed by self tuner for effective response. Since the proposed controller doesn't include integral part, it can't eliminate steady state error. So we include a gain to eliminate the steady state error.

• Journal of Industrial Technology
• /
• v.20 no.B
• /
• pp.155-162
• /
• 2000

Although Fuzzy Logic Controller(FLC) adopted three terms as input gives better performance, FLC is in general composed of two-term control because of the difficulty in the construction of fuzzy rule base. In this paper, a three-term FLC which is similar to PID control but acts as a nonlinear controller is proposed. To reduce the complexity of the rule base design and to increase efficiency. a simplified fuzzy PID control is induced from a hybrid velocity/position type PID algorithm by sharing a common rule base for both fuzzy PI and fuzzy PD parts. It is simple in structure, easy in implementation, and fast in calculation. The phase plane technique is applied to obtain the rule base for fuzzy two-term control and the resultant rule base is Macvicar-Whelan type. And the membership function is a Gaussian function. The frequency response information is used in tuning of the membership functions. Also a tuning strategy for the scaling factors is proposed based on the relationship between PID gain and the scaling factors. Simulation results show better performance and the effectiveness of the proposed method.

• Proceedings of the KIEE Conference
• /
• 1999.07b
• /
• pp.504-506
• /
• 1999

In this paper, we propose a method which control parametric uncertainty systems using PID controller by fuzzy auto tuning. We get the error and the error change rate of plant output correspond to the initial value of parameter using the Ziegler-Nickols tuning and determine the new proportional gain$(K_p)$ and the integral time $(T_i)$ from fuzzy tuner by the error and error change rate of plant output as a membership function of fuzzy theory. The Fuzzy Auto-tuning algorithm for PID controller operate to adapt variable parameter of plant in parametric uncertainty systems. It is shown this method considerably improve the transient response at computer simulation.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.06a
• /
• pp.701-706
• /
• 1998

A design method of rule -based fuzzy modeling is presented for the model identification of complex and nonlinear systems. The proposed rule-based fuzzy modeling implements system structure and parameter identification in the efficient form of " IF..., THEN,," statements. using the theories of optimization and linguistic fuzzy implication rules. The improved complex method, which is a powerful auto-tuning algorithm, is used for tuning of parameters of the premise membership functions in consideration of the overall structure of fuzzy rules. The optimized objective function, including the weighting factors, is auto-tuned for better performance of fuzzy model using training data and testing data. According to the adjustment of each weighting factor of training and testing data, we can construct the optimal fuzzy model from the objective function. The least square method is utilized for the identification of optimum consequence parameters. Gas furance and a sewage treatment proce s are used to evaluate the performance of the proposed rule-based fuzzy modeling.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.2908-2910
• /
• 1999

In this paper, a fuzzy model is identified and optimized using the hybrid algorithm and HCM clustering method. Here, the hybrid algorithm is carried out as the structure combined with both a genetic algorithm and the improved complex method. The one is utilized for determining the initial parameters of membership function, the other for obtaining the fine parameters of membership function. HCM clustering algorithm is used to determine the confined region of initial parameters and also to avoid overflow phenomenon during auto-tuning of hybrid algorithm. And the standard least square method is used for the identification of optimum consequence parameters of fuzzy model. Two numerical examples are shown to evaluate the performance of the proposed model.